Release: v3.1.0

* fix generate for GPT2 Double Head

* fix gpt2 double head model

* fix  bart / t5

* also add for no beam search

* fix no beam search

* fix encoder decoder

* simplify t5

* simplify t5

* fix t5 tests

* fix BART

* fix transfo-xl

* fix conflict

* integrating sylvains and sams comments

* fix tf past_decoder_key_values

* fix enc dec test

.circleci/config.yml

@@ -1,4 +1,66 @@
-version: 2
+version: 2.1
+orbs:
+    gcp-gke: circleci/gcp-gke@1.0.4
+    go: circleci/go@1.3.0
+
+# TPU REFERENCES
+references:
+    checkout_ml_testing: &checkout_ml_testing
+        run:
+            name: Checkout ml-testing-accelerators
+            command: |
+                git clone https://github.com/GoogleCloudPlatform/ml-testing-accelerators.git
+                cd ml-testing-accelerators
+                git fetch origin 5e88ac24f631c27045e62f0e8d5dfcf34e425e25:stable
+                git checkout stable
+    build_push_docker: &build_push_docker
+        run:
+            name: Configure Docker
+            command: |
+                gcloud --quiet auth configure-docker
+                cd docker/transformers-pytorch-tpu
+                if [ -z "$CIRCLE_PR_NUMBER" ]; then docker build --tag "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID" -f Dockerfile --build-arg "TEST_IMAGE=1" . ; else docker build --tag "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID" -f Dockerfile --build-arg "TEST_IMAGE=1" --build-arg "GITHUB_REF=pull/$CIRCLE_PR_NUMBER/head" . ; fi
+                docker push "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID"
+    deploy_cluster: &deploy_cluster
+        run:
+            name: Deploy the job on the kubernetes cluster
+            command: |
+                go get github.com/google/go-jsonnet/cmd/jsonnet && \
+                export PATH=$PATH:$HOME/go/bin && \
+                kubectl create -f docker/transformers-pytorch-tpu/dataset.yaml || true && \
+                job_name=$(jsonnet -J ml-testing-accelerators/ docker/transformers-pytorch-tpu/bert-base-cased.jsonnet --ext-str image=$GCR_IMAGE_PATH --ext-str image-tag=$CIRCLE_WORKFLOW_JOB_ID | kubectl create -f -) && \
+                job_name=${job_name#job.batch/} && \
+                job_name=${job_name% created} && \
+                echo "Waiting on kubernetes job: $job_name" && \
+                i=0 && \
+                # 30 checks spaced 30s apart = 900s total.
+                max_checks=30 && \
+                status_code=2 && \
+                # Check on the job periodically. Set the status code depending on what
+                # happened to the job in Kubernetes. If we try max_checks times and
+                # still the job hasn't finished, give up and return the starting
+                # non-zero status code.
+                while [ $i -lt $max_checks ]; do ((i++)); if kubectl get jobs $job_name -o jsonpath='Failed:{.status.failed}' | grep "Failed:1"; then status_code=1 && break; elif kubectl get jobs $job_name -o jsonpath='Succeeded:{.status.succeeded}' | grep "Succeeded:1" ; then status_code=0 && break; else echo "Job not finished yet"; fi; sleep 30; done && \
+                echo "Done waiting. Job status code: $status_code" && \
+                pod_name=$(kubectl get po -l controller-uid=`kubectl get job $job_name -o "jsonpath={.metadata.labels.controller-uid}"` | awk 'match($0,!/NAME/) {print $1}') && \
+                echo "GKE pod name: $pod_name" && \
+                kubectl logs -f $pod_name --container=train
+                echo "Done with log retrieval attempt." && \
+                gcloud container images delete "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID" --force-delete-tags && \
+                exit $status_code
+    delete_gke_jobs: &delete_gke_jobs
+        run:
+            name: Delete GKE Jobs
+            command: |
+                # Match jobs whose age matches patterns like '1h' or '1d', i.e. any job
+                # that has been around longer than 1hr. First print all columns for
+                # matches, then execute the delete.
+                kubectl get job | awk 'match($4,/[0-9]+[dh]/) {print $0}'
+                kubectl delete job $(kubectl get job | awk 'match($4,/[0-9]+[dh]/) {print $1}')
+
+
+
+
 jobs:
     run_tests_torch_and_tf:
         working_directory: ~/transformers
@@ -10,9 +72,19 @@ jobs:
         parallelism: 1
         steps:
             - checkout
-            - run: sudo pip install .[sklearn,tf-cpu,torch,testing]
-            - run: sudo pip install codecov pytest-cov
-            - run: python -m pytest -n 8 --dist=loadfile -s ./tests/ --cov  | tee output.txt
+            - restore_cache:
+                  keys:
+                      - v0.3-torch_and_tf-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install git+https://github.com/huggingface/nlp
+            - run: pip install .[sklearn,tf-cpu,torch,testing]
+            - run: pip install codecov pytest-cov
+            - save_cache:
+                key: v0.3-{{ checksum "setup.py" }}
+                paths:
+                    - '~/.cache/pip'
+            - run: python -m pytest -n 8 --dist=loadfile -rA -s ./tests/ --cov  | tee output.txt
             - run: codecov
             - store_artifacts:
                   path: ~/transformers/output.txt
@@ -27,12 +99,21 @@ jobs:
         parallelism: 1
         steps:
             - checkout
-            - run: sudo pip install .[sklearn,torch,testing]
-            - run: python -m pytest -n 8 --dist=loadfile -s ./tests/ | tee output.txt
+            - restore_cache:
+                  keys:
+                      - v0.3-torch-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install git+https://github.com/huggingface/nlp
+            - run: pip install .[sklearn,torch,testing]
+            - save_cache:
+                  key: v0.3-torch-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
+            - run: python -m pytest -n 8 --dist=loadfile -rA -s ./tests/ | tee output.txt
             - store_artifacts:
                   path: ~/transformers/output.txt
                   destination: test_output.txt
-
     run_tests_tf:
         working_directory: ~/transformers
         docker:
@@ -43,8 +124,18 @@ jobs:
         parallelism: 1
         steps:
             - checkout
-            - run: sudo pip install .[sklearn,tf-cpu,testing]
-            - run: python -m pytest -n 8 --dist=loadfile -s ./tests/ | tee output.txt
+            - restore_cache:
+                  keys:
+                      - v0.3-tf-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install git+https://github.com/huggingface/nlp
+            - run: pip install .[sklearn,tf-cpu,testing]
+            - save_cache:
+                  key: v0.3-tf-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
+            - run: python -m pytest -n 8 --dist=loadfile -rA -s ./tests/ | tee output.txt
             - store_artifacts:
                path: ~/transformers/output.txt
                destination: test_output.txt
@@ -56,8 +147,21 @@ jobs:
             RUN_CUSTOM_TOKENIZERS: yes
         steps:
             - checkout
-            - run: sudo pip install .[mecab,testing]
-            - run: python -m pytest -sv ./tests/test_tokenization_bert_japanese.py
+            - restore_cache:
+                  keys:
+                      - v0.3-custom_tokenizers-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install .[ja,testing]
+            - run: python -m unidic download
+            - save_cache:
+                  key: v0.3-custom_tokenizers-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
+            - run: python -m pytest -s ./tests/test_tokenization_bert_japanese.py | tee output.txt
+            - store_artifacts:
+                path: ~/transformers/output.txt
+                destination: test_output.txt
     run_examples_torch:
         working_directory: ~/transformers
         docker:
@@ -68,9 +172,18 @@ jobs:
         parallelism: 1
         steps:
             - checkout
-            - run: sudo pip install .[sklearn,torch,testing]
-            - run: sudo pip install -r examples/requirements.txt
-            - run: python -m pytest -n 8 --dist=loadfile -s ./examples/ | tee output.txt
+            - restore_cache:
+                  keys:
+                      - v0.3-torch_examples-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install .[sklearn,torch,testing]
+            - run: pip install -r examples/requirements.txt
+            - save_cache:
+                  key: v0.3-torch_examples-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
+            - run: python -m pytest -n 8 --dist=loadfile -rA -s ./examples/ | tee output.txt
             - store_artifacts:
                   path: ~/transformers/output.txt
                   destination: test_output.txt
@@ -80,7 +193,16 @@ jobs:
             - image: circleci/python:3.6
         steps:
             - checkout
-            - run: sudo pip install .[tf,torch,docs]
+            - restore_cache:
+                  keys:
+                      - v0.3-build_doc-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install .[tf,torch,docs]
+            - save_cache:
+                  key: v0.3-build_doc-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
             - run: cd docs && make html SPHINXOPTS="-W"
             - store_artifacts:
                 path: ./docs/_build
@@ -93,7 +215,15 @@ jobs:
                 fingerprints:
                     - "5b:7a:95:18:07:8c:aa:76:4c:60:35:88:ad:60:56:71"
             - checkout
-            - run: sudo pip install .[tf,torch,docs]
+            - restore_cache:
+                  keys:
+                      - v0.3-deploy_doc-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install .[tf,torch,docs]
+            - save_cache:
+                  key: v0.3-deploy_doc-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
             - run: ./.circleci/deploy.sh
     check_code_quality:
         working_directory: ~/transformers
@@ -103,12 +233,21 @@ jobs:
         parallelism: 1
         steps:
             - checkout
-            # we need a version of isort with https://github.com/timothycrosley/isort/pull/1000
-            - run: sudo pip install git+git://github.com/timothycrosley/isort.git@e63ae06ec7d70b06df9e528357650281a3d3ec22#egg=isort
-            - run: sudo pip install .[tf,torch,quality]
+            - restore_cache:
+                  keys:
+                      - v0.3-code_quality-{{ checksum "setup.py" }}
+                      - v0.3-{{ checksum "setup.py" }}
+            - run: pip install --upgrade pip
+            - run: pip install isort
+            - run: pip install .[tf,torch,quality]
+            - save_cache:
+                  key: v0.3-code_quality-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
             - run: black --check --line-length 119 --target-version py35 examples templates tests src utils
             - run: isort --check-only --recursive examples templates tests src utils
             - run: flake8 examples templates tests src utils
+            - run: python utils/check_repo.py
     check_repository_consistency:
         working_directory: ~/transformers
         docker:
@@ -117,8 +256,37 @@ jobs:
         parallelism: 1
         steps:
             - checkout
-            - run: sudo pip install requests
+            - run: pip install requests
             - run: python ./utils/link_tester.py
+
+# TPU JOBS
+    run_examples_tpu:
+        docker:
+            - image: circleci/python:3.6
+        environment:
+            OMP_NUM_THREADS: 1
+        resource_class: xlarge
+        parallelism: 1
+        steps:
+            - checkout
+            - go/install
+            - *checkout_ml_testing
+            - gcp-gke/install
+            - gcp-gke/update-kubeconfig-with-credentials:
+                  cluster: $GKE_CLUSTER
+                  perform-login: true
+            - setup_remote_docker
+            - *build_push_docker
+            - *deploy_cluster
+    cleanup-gke-jobs:
+        docker:
+            - image: circleci/python:3.6
+        steps:
+            - gcp-gke/install
+            - gcp-gke/update-kubeconfig-with-credentials:
+                  cluster: $GKE_CLUSTER
+                  perform-login: true
+            - *delete_gke_jobs
 workflow_filters: &workflow_filters
     filters:
         branches:
@@ -137,3 +305,15 @@ workflows:
             - run_tests_tf
             - build_doc
             - deploy_doc: *workflow_filters
+    tpu_testing_jobs:
+        triggers:
+            - schedule:
+                # Set to run at the first minute of every hour.
+                cron: "0 8 * * *"
+                filters:
+                    branches:
+                        only:
+                            - master
+        jobs:
+            - cleanup-gke-jobs
+            - run_examples_tpu

.circleci/deploy.sh

@@ -46,4 +46,5 @@ deploy_doc "11c3257" v2.8.0
 deploy_doc "e7cfc1a" v2.9.0
 deploy_doc "7cb203f" v2.9.1
 deploy_doc "10d7239" v2.10.0 
-deploy_doc "b42586e" #v2.11.0 Latest stable release
+deploy_doc "b42586e" v2.11.0
+deploy_doc "7fb8bdf" #v3.0.2 Latest stable release

.github/ISSUE_TEMPLATE/bug-report.md

@@ -7,14 +7,53 @@ assignees: ''
 
 ---
 
-# 🐛 Bug
+
+## Environment info
+<!-- You can run the command `transformers-cli env` and copy-and-paste its output below.
+     Don't forget to fill out the missing fields in that output! -->
+     
+- `transformers` version:
+- Platform:
+- Python version:
+- PyTorch version (GPU?):
+- Tensorflow version (GPU?):
+- Using GPU in script?:
+- Using distributed or parallel set-up in script?:
+
+### Who can help
+<!-- Your issue will be replied to more quickly if you can figure out the right person to tag with @
+ If you know how to use git blame, that is the easiest way, otherwise, here is a rough guide of **who to tag**.
+ Please tag fewer than 3 people.
+ 
+ albert, bert, GPT2, XLM: @LysandreJik 
+ tokenizers: @mfuntowicz
+ Trainer: @sgugger
+ Speed and Memory Benchmarks: @patrickvonplaten
+ Model Cards: @julien-c
+ Translation: @sshleifer
+ Summarization: @sshleifer
+ TextGeneration: @TevenLeScao 
+ examples/distillation: @VictorSanh
+ nlp datasets: [different repo](https://github.com/huggingface/nlp)
+ rust tokenizers: [different repo](https://github.com/huggingface/tokenizers)
+ Text Generation: @TevenLeScao
+ blenderbot: @mariamabarham
+ Bart: @sshleifer
+ Marian: @sshleifer
+ T5: @patrickvonplaten
+ Longformer/Reformer: @patrickvonplaten
+ TransfoXL/XLNet: @TevenLeScao 
+ examples/seq2seq: @sshleifer
+ examples/bert-loses-patience: @JetRunner
+ tensorflow: @jplu
+ examples/token-classification: @stefan-it
+ documentation: @sgugger
+ -->
 
 ## Information
 
 Model I am using (Bert, XLNet ...):
 
-Language I am using the model on (English, Chinese ...):
-
 The problem arises when using:
 * [ ] the official example scripts: (give details below)
 * [ ] my own modified scripts: (give details below)
@@ -38,15 +77,3 @@ Steps to reproduce the behavior:
 ## Expected behavior
 
 <!-- A clear and concise description of what you would expect to happen. -->
-
-## Environment info
-<!-- You can run the command `transformers-cli env` and copy-and-paste its output below.
-     Don't forget to fill out the missing fields in that output! -->
-     
-- `transformers` version:
-- Platform:
-- Python version:
-- PyTorch version (GPU?):
-- Tensorflow version (GPU?):
-- Using GPU in script?:
-- Using distributed or parallel set-up in script?:

.github/ISSUE_TEMPLATE/question-help.md

@@ -1,6 +1,6 @@
 ---
 name: "❓ Questions & Help"
-about: Post your general questions on Stack Overflow tagged huggingface-transformers
+about: Post your general questions on the Hugging Face forum or Stack Overflow tagged huggingface-transformers
 title: ''
 labels: ''
 assignees: ''
@@ -11,19 +11,17 @@ assignees: ''
 
 <!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
      new models and benchmarks, and migration questions. For all other questions,
-     we direct you to Stack Overflow (SO) where a whole community of PyTorch and
-     Tensorflow enthusiast can help you out. Make sure to tag your question with the
-     right deep learning framework as well as the huggingface-transformers tag: 
+     we direct you to the Hugging Face forum: https://discuss.huggingface.co/ .
+     You can also try Stack Overflow (SO) where a whole community of PyTorch and
+     Tensorflow enthusiast can help you out. In this case, make sure to tag your
+     question with the right deep learning framework as well as the
+     huggingface-transformers tag: 
      https://stackoverflow.com/questions/tagged/huggingface-transformers 
-     
-     If your question wasn't answered after a period of time on Stack Overflow, you
-     can always open a question on GitHub. You should then link to the SO question 
-     that you posted.
      -->
 
 ## Details
 <!-- Description of your issue -->
 
-<!-- You should first ask your question on SO, and only if
+<!-- You should first ask your question on the forum or SO, and only if
      you didn't get an answer ask it here on GitHub. -->
-**A link to original question on Stack Overflow**:
+**A link to original question on the forum/Stack Overflow**:

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,2 @@
+<!-- This line specifies which issue to close after the pull request is merged. -->
+Fixes #{issue number}

.github/workflows/github-push.yml

@@ -1,19 +0,0 @@
-name: GitHub-hosted runner
-
-on: push
-
-jobs:
-  check_code_quality:
-    runs-on: ubuntu-18.04
-    steps:
-    - uses: actions/checkout@v2
-    - name: Set up Python
-      uses: actions/setup-python@v1
-      with:
-        python-version: 3.7
-    # - name: Install dependencies
-    #   run: |
-    #     pip install .[tf,torch,quality]
-
-
-

.github/workflows/github-torch-hub.yml

@@ -18,8 +18,17 @@ jobs:
       uses: actions/setup-python@v1
       with:
         python-version: 3.7
+
+    - name: Loading cache
+      uses: actions/cache@v2
+      id: cache
+      with:
+        path: ~/.cache/pip
+        key: v0-torch_hub-${{ hashFiles('setup.py') }}
+
     - name: Install dependencies
       run: |
+        pip install --upgrade pip
         pip install torch
         pip install numpy tokenizers filelock requests tqdm regex sentencepiece sacremoses packaging
 

.github/workflows/self-push.yml

@@ -25,6 +25,14 @@ jobs:
     - name: Current dir
       run: pwd
     - run: nvidia-smi
+
+    - name: Loading cache.
+      uses: actions/cache@v2
+      id: cache
+      with:
+        path: .env
+        key: v0-tests_tf_torch_gpu-${{ hashFiles('setup.py') }}
+
     - name: Create new python env (on self-hosted runners we have to handle isolation ourselves)
       run: |
         python -m venv .env
@@ -35,8 +43,10 @@ jobs:
     - name: Install dependencies
       run: |
         source .env/bin/activate
-        pip install torch
-        pip install .[sklearn,testing]
+        pip install --upgrade pip
+        pip install torch!=1.6.0
+        pip install .[sklearn,testing,onnxruntime]
+        pip install git+https://github.com/huggingface/nlp
 
     - name: Are GPUs recognized by our DL frameworks
       run: |
@@ -51,4 +61,4 @@ jobs:
         USE_CUDA: yes
       run: |
         source .env/bin/activate
-        python -m pytest -n 2 --dist=loadfile -s -v ./tests/
+        python -m pytest -n 2 --dist=loadfile -s ./tests/

.github/workflows/self-scheduled.yml

@@ -13,6 +13,14 @@ jobs:
     runs-on: self-hosted
     steps:
     - uses: actions/checkout@v2
+
+    - name: Loading cache.
+      uses: actions/cache@v2
+      id: cache
+      with:
+        path: .env
+        key: v0-slow_tests_tf_torch_gpu-${{ hashFiles('setup.py') }}
+
     - name: Python version
       run: |
         which python
@@ -22,6 +30,7 @@ jobs:
       run: pwd
     - run: nvidia-smi
     - name: Create new python env (on self-hosted runners we have to handle isolation ourselves)
+      if: steps.cache.outputs.cache-hit != 'true'
       run: |
         python -m venv .env
         source .env/bin/activate
@@ -31,7 +40,10 @@ jobs:
     - name: Install dependencies
       run: |
         source .env/bin/activate
-        pip install .[sklearn,torch,testing]
+        pip install --upgrade pip
+        pip install torch!=1.6.0
+        pip install .[sklearn,testing,onnxruntime]
+        pip install git+https://github.com/huggingface/nlp
 
     - name: Are GPUs recognized by our DL frameworks
       run: |
@@ -46,5 +58,15 @@ jobs:
         USE_CUDA: yes
       run: |
         source .env/bin/activate
-        python -m pytest -n 1 --dist=loadfile -s -v ./tests/
-        
+        python -m pytest -n 1 --dist=loadfile -s ./tests/
+
+    - name: Run examples tests on GPU
+      env:
+        TF_FORCE_GPU_ALLOW_GROWTH: "true"
+        OMP_NUM_THREADS: 1
+        RUN_SLOW: yes
+        USE_CUDA: yes
+      run: |
+        source .env/bin/activate
+        pip install -r examples/requirements.txt
+        python -m pytest -n 1 --dist=loadfile -s examples

CONTRIBUTING.md

@@ -65,8 +65,8 @@ Awesome! Please provide the following information:
 If you are willing to contribute the model yourself, let us know so we can best
 guide you.
 
-We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them 
-in the [`templates`](https://github.com/huggingface/transformers/templates) folder.
+We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them
+in the [`templates`](https://github.com/huggingface/transformers/tree/master/templates) folder.
 
 ### Do you want a new feature (that is not a model)?
 
@@ -87,8 +87,8 @@ A world-class feature request addresses the following points:
 If your issue is well written we're already 80% of the way there by the time you
 post it.
 
-We have added **templates** to guide you in the process of adding a new example script for training or testing the 
-models in the library. You can find them in the [`templates`](https://github.com/huggingface/transformers/templates) 
+We have added **templates** to guide you in the process of adding a new example script for training or testing the
+models in the library. You can find them in the [`templates`](https://github.com/huggingface/transformers/tree/master/templates)
 folder.
 
 ## Start contributing! (Pull Requests)
@@ -134,12 +134,6 @@ Follow these steps to start contributing:
    it with `pip uninstall transformers` before reinstalling it in editable
    mode with the `-e` flag.)
 
-   Right now, we need an unreleased version of `isort` to avoid a
-   [bug](https://github.com/timothycrosley/isort/pull/1000):
-
-   ```bash
-   $ pip install -U git+git://github.com/timothycrosley/isort.git@e63ae06ec7d70b06df9e528357650281a3d3ec22#egg=isort
-   ```
 5. Develop the features on your branch.
 
    As you work on the features, you should make sure that the test suite
@@ -149,6 +143,14 @@ Follow these steps to start contributing:
    $ make test
    ```
 
+   Note, that this command uses `-n auto` pytest flag, therefore, it will start as many parallel `pytest` processes as the number of your computer's CPU-cores, and if you have lots of those and a few GPUs and not a great amount of RAM, it's likely to overload your computer. Therefore, to run the test suite, you may want to consider using this command instead:
+
+   ```bash
+   $ python -m pytest -n 3 --dist=loadfile -s -v ./tests/
+   ```
+
+   Adjust the value of `-n` to fit the load your hardware can support.
+
    `transformers` relies on `black` and `isort` to format its source code
    consistently. After you make changes, format them with:
 
@@ -163,6 +165,16 @@ Follow these steps to start contributing:
    $ make quality
    ```
 
+   If you're modifying documents under `docs/source`, make sure to validate that
+   they can still be built. This check also runs in CI. To run a local check
+   make sure you have installed the documentation builder requirements, by
+   running `pip install .[tf,torch,docs]` once from the root of this repository
+   and then run:
+
+   ```bash
+   $ make docs
+   ```
+
    Once you're happy with your changes, add changed files using `git add` and
    make a commit with `git commit` to record your changes locally:
 
@@ -208,21 +220,21 @@ Follow these steps to start contributing:
    are useful to avoid duplicated work, and to differentiate it from PRs ready
    to be merged;
 4. Make sure existing tests pass;
-5. Add high-coverage tests. No quality testing = no merge. 
-   - If you are adding a new model, make sure that you use 
+5. Add high-coverage tests. No quality testing = no merge.
+   - If you are adding a new model, make sure that you use
      `ModelTester.all_model_classes = (MyModel, MyModelWithLMHead,...)`, which triggers the common tests.
-   - If you are adding new `@slow` tests, make sure they pass using 
-     `RUN_SLOW=1 python -m pytest tests/test_my_new_model.py`. 
-   - If you are adding a new tokenizer, write tests, and make sure 
+   - If you are adding new `@slow` tests, make sure they pass using
+     `RUN_SLOW=1 python -m pytest tests/test_my_new_model.py`.
+   - If you are adding a new tokenizer, write tests, and make sure
      `RUN_SLOW=1 python -m pytest tests/test_tokenization_{your_model_name}.py` passes.
-   CircleCI does not run the slow tests. 
-6. All public methods must have informative docstrings that work nicely with sphinx. See `modeling_ctrl.py` for an 
+   CircleCI does not run the slow tests, but github actions does every night!
+6. All public methods must have informative docstrings that work nicely with sphinx. See `modeling_ctrl.py` for an
    example.
 
 ### Tests
 
-An extensive test suite is included to test the library behavior and several examples. Library tests can be found in 
-the [tests folder](https://github.com/huggingface/transformers/tree/master/tests) and examples tests in the 
+An extensive test suite is included to test the library behavior and several examples. Library tests can be found in
+the [tests folder](https://github.com/huggingface/transformers/tree/master/tests) and examples tests in the
 [examples folder](https://github.com/huggingface/transformers/tree/master/examples).
 
 We like `pytest` and `pytest-xdist` because it's faster. From the root of the
@@ -238,8 +250,7 @@ and for the examples:
 $ pip install -r examples/requirements.txt  # only needed the first time
 $ python -m pytest -n auto --dist=loadfile -s -v ./examples/
 ```
-
-In fact, that's how `make test` and `make test-examples` are implemented!
+In fact, that's how `make test` and `make test-examples` are implemented (sans the `pip install` line)!
 
 You can specify a smaller set of tests in order to test only the feature
 you're working on.

Makefile

@@ -1,17 +1,18 @@
-.PHONY: quality style test test-examples
+.PHONY: quality style test test-examples docs
 
 # Check that source code meets quality standards
 
 quality:
 	black --check --line-length 119 --target-version py35 examples templates tests src utils
-	isort --check-only --recursive examples templates tests src utils
+	isort --check-only examples templates tests src utils
 	flake8 examples templates tests src utils
+	python utils/check_repo.py
 
 # Format source code automatically
 
 style:
 	black --line-length 119 --target-version py35 examples templates tests src utils
-	isort --recursive examples templates tests src utils
+	isort examples templates tests src utils
 
 # Run tests for the library
 
@@ -22,3 +23,8 @@ test:
 
 test-examples:
 	python -m pytest -n auto --dist=loadfile -s -v ./examples/
+
+# Check that docs can build
+
+docs:
+	cd docs && make html SPHINXOPTS="-W"

README.md

@@ -167,19 +167,31 @@ At some point in the future, you'll be able to seamlessly move from pre-training
 19. **[Reformer](https://huggingface.co/transformers/model_doc/reformer.html)** (from Google Research) released with the paper [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451) by Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya.
 20. **[MarianMT](https://huggingface.co/transformers/model_doc/marian.html)** Machine translation models trained using [OPUS](http://opus.nlpl.eu/) data by Jörg Tiedemann. The [Marian Framework](https://marian-nmt.github.io/) is being developed by the Microsoft Translator Team.
 21. **[Longformer](https://huggingface.co/transformers/model_doc/longformer.html)** (from AllenAI) released with the paper [Longformer: The Long-Document Transformer](https://arxiv.org/abs/2004.05150) by Iz Beltagy, Matthew E. Peters, Arman Cohan.
-22. **[Other community models](https://huggingface.co/models)**, contributed by the [community](https://huggingface.co/users).
-23. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
+22. **[DPR](https://github.com/facebookresearch/DPR)** (from Facebook) released with the paper [Dense Passage Retrieval
+for Open-Domain Question Answering](https://arxiv.org/abs/2004.04906) by Vladimir Karpukhin, Barlas Oğuz, Sewon
+Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih.
+23. **[Pegasus](https://github.com/google-research/pegasus)** (from Google) released with the paper [PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization](https://arxiv.org/abs/1912.08777)> by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu.
+24. **[MBart](https://github.com/pytorch/fairseq/tree/master/examples/mbart)** (from Facebook) released with the paper  [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer.  
+25. **[Other community models](https://huggingface.co/models)**, contributed by the [community](https://huggingface.co/users).
+26. Want to contribute a new model? We have added a **detailed guide and templates** to guide you in the process of adding a new model. You can find them in the [`templates`](./templates) folder of the repository. Be sure to check the [contributing guidelines](./CONTRIBUTING.md) and contact the maintainers or open an issue to collect feedbacks before starting your PR.
 
 These implementations have been tested on several datasets (see the example scripts) and should match the performances of the original implementations (e.g. ~93 F1 on SQuAD for BERT Whole-Word-Masking, ~88 F1 on RocStories for OpenAI GPT, ~18.3 perplexity on WikiText 103 for Transformer-XL, ~0.916 Pearson R coefficient on STS-B for XLNet). You can find more details on the performances in the Examples section of the [documentation](https://huggingface.co/transformers/examples.html).
 
 ## Online demo
 
+You can test our inference API on most model pages from the model hub: https://huggingface.co/models
+
+For example: 
+- [Masked word completion with BERT](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
+- [NER with Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
+- [Text generation with GPT-2](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
+- [NLI with RoBERTa](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
+- [Summarization with BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
+- [Question answering with DistilBERT](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
+- [Translation with T5](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
+
+
 **[Write With Transformer](https://transformer.huggingface.co)**, built by the Hugging Face team at transformer.huggingface.co, is the official demo of this repo’s text generation capabilities.
-You can use it to experiment with completions generated by `GPT2Model`, `TransfoXLModel`, and `XLNetModel`.
-
-> “🦄 Write with transformer is to writing what calculators are to calculus.”
-
-![write_with_transformer](https://transformer.huggingface.co/front/assets/thumbnail-large.png)
 
 ## Quick tour
 
@@ -613,7 +625,7 @@ Breaking change in the `from_pretrained()` method:
 
 1. Models are now set in evaluation mode by default when instantiated with the `from_pretrained()` method. To train them, don't forget to set them back in training mode (`model.train()`) to activate the dropout modules.
 
-2. The additional `*input` and `**kwargs` arguments supplied to the `from_pretrained()` method used to be directly passed to the underlying model's class `__init__()` method. They are now used to update the model configuration attribute instead, which can break derived model classes built based on the previous `BertForSequenceClassification` examples. We are working on a way to mitigate this breaking change in [#866](https://github.com/huggingface/transformers/pull/866) by forwarding the the model's `__init__()` method (i) the provided positional arguments and (ii) the keyword arguments which do not match any configuration class attributes.
+2. The additional `*input` and `**kwargs` arguments supplied to the `from_pretrained()` method used to be directly passed to the underlying model's class `__init__()` method. They are now used to update the model configuration attribute instead, which can break derived model classes built based on the previous `BertForSequenceClassification` examples. We are working on a way to mitigate this breaking change in [#866](https://github.com/huggingface/transformers/pull/866) by forwarding the model's `__init__()` method (i) the provided positional arguments and (ii) the keyword arguments which do not match any configuration class attributes.
 
 Also, while not a breaking change, the serialization methods have been standardized and you probably should switch to the new method `save_pretrained(save_directory)` if you were using any other serialization method before.
 
@@ -686,11 +698,11 @@ for batch in train_data:
 
 ## Citation
 
-We now have a paper you can cite for the 🤗 Transformers library:
+We now have a [paper](https://arxiv.org/abs/1910.03771) you can cite for the 🤗 Transformers library:
 ```bibtex
 @article{Wolf2019HuggingFacesTS,
   title={HuggingFace's Transformers: State-of-the-art Natural Language Processing},
-  author={Thomas Wolf and Lysandre Debut and Victor Sanh and Julien Chaumond and Clement Delangue and Anthony Moi and Pierric Cistac and Tim Rault and R'emi Louf and Morgan Funtowicz and Jamie Brew},
+  author={Thomas Wolf and Lysandre Debut and Victor Sanh and Julien Chaumond and Clement Delangue and Anthony Moi and Pierric Cistac and Tim Rault and Rémi Louf and Morgan Funtowicz and Joe Davison and Sam Shleifer and Patrick von Platen and Clara Ma and Yacine Jernite and Julien Plu and Canwen Xu and Teven Le Scao and Sylvain Gugger and Mariama Drame and Quentin Lhoest and Alexander M. Rush},
   journal={ArXiv},
   year={2019},
   volume={abs/1910.03771}

codecov.yml

@@ -4,3 +4,7 @@ coverage:
       default:
         informational: true
     patch: off
+comment:
+  require_changes: true    # only comment if there was change in coverage
+  require_head: yes        # don't report if there is no head coverage report
+  require_base: yes        # don't report if there is no base coverage report

deploy_multi_version_doc.sh

@@ -1,23 +0,0 @@
-cd docs
-
-function deploy_doc(){
-	echo "Creating doc at commit $1 and pushing to folder $2"
-	git checkout $1
-	if [ ! -z "$2" ] 
-	then
-		echo "Pushing version" $2
-		make clean && make html && scp -r -oStrictHostKeyChecking=no _build/html $doc:$dir/$2
-	else
-		echo "Pushing master"
-		make clean && make html && scp -r -oStrictHostKeyChecking=no _build/html/* $doc:$dir
-	fi
-}
-
-deploy_doc "master" 
-deploy_doc "b33a385" v1.0.0
-deploy_doc "fe02e45" v1.1.0
-deploy_doc "89fd345" v1.2.0
-deploy_doc "fc9faa8" v2.0.0
-deploy_doc "3ddce1d" v2.1.1
-deploy_doc "f2f3294" v2.2.0
-deploy_doc "d0f8b9a" v2.3.0

docker/transformers-pytorch-tpu/Dockerfile

@@ -0,0 +1,65 @@
+FROM google/cloud-sdk:slim
+
+# Build args.
+ARG GITHUB_REF=refs/heads/master
+
+# TODO: This Dockerfile installs pytorch/xla 3.6 wheels. There are also 3.7
+# wheels available; see below.
+ENV PYTHON_VERSION=3.6
+
+RUN apt-get update && apt-get install -y --no-install-recommends \
+         build-essential \
+         cmake \
+         git \
+         curl \
+         ca-certificates
+
+# Install conda and python.
+# NOTE new Conda does not forward the exit status... https://github.com/conda/conda/issues/8385
+RUN curl -o ~/miniconda.sh https://repo.anaconda.com/miniconda/Miniconda3-4.7.12-Linux-x86_64.sh  && \
+    chmod +x ~/miniconda.sh && \
+    ~/miniconda.sh -b && \
+    rm ~/miniconda.sh
+
+ENV PATH=/root/miniconda3/bin:$PATH
+
+RUN conda create -y --name container python=$PYTHON_VERSION
+
+# Run the rest of commands within the new conda env.
+# Use absolute path to appease Codefactor.
+SHELL ["/root/miniconda3/bin/conda", "run", "-n", "container", "/bin/bash", "-c"]
+RUN conda install -y python=$PYTHON_VERSION mkl
+
+RUN pip uninstall -y torch && \
+    # Python 3.7 wheels are available. Replace cp36-cp36m with cp37-cp37m
+    gsutil cp 'gs://tpu-pytorch/wheels/torch-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' . && \
+    gsutil cp 'gs://tpu-pytorch/wheels/torch_xla-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' . && \
+    gsutil cp 'gs://tpu-pytorch/wheels/torchvision-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' . && \
+    pip install 'torch-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' && \
+    pip install 'torch_xla-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' && \
+    pip install 'torchvision-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' && \
+    rm 'torch-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' && \
+    rm 'torch_xla-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' && \
+    rm 'torchvision-nightly-cp${PYTHON_VERSION/./}-cp${PYTHON_VERSION/./}m-linux_x86_64.whl' && \
+    apt-get install -y libomp5
+
+ENV LD_LIBRARY_PATH=root/miniconda3/envs/container/lib
+
+
+# Install huggingface/transformers at the current PR, plus dependencies.
+RUN git clone https://github.com/huggingface/transformers.git && \
+    cd transformers && \
+    git fetch origin $GITHUB_REF:CI && \
+    git checkout CI && \
+    cd .. && \
+    pip install ./transformers && \
+    pip install -r ./transformers/examples/requirements.txt && \
+    pip install pytest
+
+RUN python -c "import torch_xla; print(torch_xla.__version__)"
+RUN python -c "import transformers as trf; print(trf.__version__)"
+RUN conda init bash
+COPY docker-entrypoint.sh /usr/local/bin/
+RUN chmod +x /usr/local/bin/docker-entrypoint.sh
+ENTRYPOINT ["/usr/local/bin/docker-entrypoint.sh"]
+CMD ["bash"]

docker/transformers-pytorch-tpu/bert-base-cased.jsonnet

@@ -0,0 +1,38 @@
+local base = import 'templates/base.libsonnet';
+local tpus = import 'templates/tpus.libsonnet';
+local utils = import "templates/utils.libsonnet";
+local volumes = import "templates/volumes.libsonnet";
+
+local bertBaseCased = base.BaseTest {
+  frameworkPrefix: "hf",
+  modelName: "bert-base-cased",
+  mode: "example",
+  configMaps: [],
+
+  timeout: 3600, # 1 hour, in seconds
+
+  image: std.extVar('image'),
+  imageTag: std.extVar('image-tag'),
+
+  tpuSettings+: {
+    softwareVersion: "pytorch-nightly",
+  },
+  accelerator: tpus.v3_8,
+
+  volumeMap+: {
+    datasets: volumes.PersistentVolumeSpec {
+      name: "huggingface-cluster-disk",
+      mountPath: "/datasets",
+    },
+  },
+  command: utils.scriptCommand(
+    |||
+      python -m pytest -s transformers/examples/test_xla_examples.py -v
+      test_exit_code=$?
+      echo "\nFinished running commands.\n"
+      test $test_exit_code -eq 0
+    |||
+  ),
+};
+
+bertBaseCased.oneshotJob

docker/transformers-pytorch-tpu/dataset.yaml

@@ -0,0 +1,32 @@
+apiVersion: v1
+kind: PersistentVolume
+metadata:
+  name: huggingface-cluster-disk
+spec:
+  storageClassName: ""
+  capacity:
+    storage: 500Gi
+  accessModes:
+    - ReadOnlyMany
+  claimRef:
+    namespace: default
+    name: huggingface-cluster-disk-claim
+  gcePersistentDisk:
+    pdName: huggingface-cluster-disk
+    fsType: ext4
+    readOnly: true
+---
+apiVersion: v1
+kind: PersistentVolumeClaim
+metadata:
+  name: huggingface-cluster-disk-claim
+spec:
+  # Specify "" as the storageClassName so it matches the PersistentVolume's StorageClass.
+  # A nil storageClassName value uses the default StorageClass. For details, see
+  # https://kubernetes.io/docs/concepts/storage/persistent-volumes/#class-1
+  storageClassName: ""
+  accessModes:
+    - ReadOnlyMany
+  resources:
+    requests:
+      storage: 1Ki

docker/transformers-pytorch-tpu/docker-entrypoint.sh

@@ -0,0 +1,8 @@
+#!/bin/bash
+source ~/.bashrc
+echo "running docker-entrypoint.sh"
+conda activate container
+echo $KUBE_GOOGLE_CLOUD_TPU_ENDPOINTS
+echo "printed TPU info"
+export XRT_TPU_CONFIG="tpu_worker;0;${KUBE_GOOGLE_CLOUD_TPU_ENDPOINTS:7}"
+exec "$@"#!/bin/bash

docs/source/_static/css/huggingface.css

@@ -1,5 +1,36 @@
 /* Our DOM objects */
 
+/* Colab dropdown */
+
+.colab-dropdown {
+    position: relative;
+    display: inline-block;
+}
+  
+.colab-dropdown-content {
+    display: none;
+    position: absolute;
+    background-color: #f9f9f9;
+    min-width: 117px;
+    box-shadow: 0px 8px 16px 0px rgba(0,0,0,0.2);
+    z-index: 1;
+}
+  
+.colab-dropdown-content button {
+    color: #6670FF;
+    background-color: #f9f9f9;
+    font-size: 12px;
+    border: none;
+    min-width: 117px;
+    padding: 5px 5px;
+    text-decoration: none;
+    display: block;
+}
+  
+.colab-dropdown-content button:hover {background-color: #eee;}
+  
+.colab-dropdown:hover .colab-dropdown-content {display: block;}
+
 /* Version control */
 
 .version-button {

docs/source/_static/js/custom.js

@@ -1,10 +1,11 @@
 // These two things need to be updated at each release for the version selector.
 // Last stable version
-const stableVersion = "v2.11.0"
+const stableVersion = "v3.0.2"
 // Dictionary doc folder to label
 const versionMapping = {
     "master": "master",
-    "": "v2.11.0 (stable)",
+    "": "v3.0.0/v3.0.1/v3.0.2 (stable)",
+    "v2.11.0": "v2.11.0",
     "v2.10.0": "v2.10.0",
     "v2.9.1": "v2.9.0/v2.9.1",
     "v2.8.0": "v2.8.0",
@@ -20,6 +21,18 @@ const versionMapping = {
     "v1.1.0": "v1.1.0",
     "v1.0.0": "v1.0.0"
 }
+// The page that have a notebook and therefore should have the open in colab badge.
+const hasNotebook = [
+    "benchmarks",
+    "custom_datasets",
+    "multilingual",
+    "perplexity",
+    "preprocessing",
+    "quicktour",
+    "task_summary",
+    "tokenizer_summary",
+    "training"
+];
 
 function addIcon() {
     const huggingFaceLogo = "https://huggingface.co/landing/assets/transformers-docs/huggingface_logo.svg";
@@ -81,12 +94,32 @@ function addGithubButton() {
     document.querySelector(".wy-side-nav-search .icon-home").insertAdjacentHTML('afterend', div);
 }
 
+function addColabLink() {
+    const parts = location.toString().split('/');
+    const pageName = parts[parts.length - 1].split(".")[0];
+
+    if (hasNotebook.includes(pageName)) {
+        const baseURL = "https://colab.research.google.com/github/huggingface/notebooks/blob/master/transformers_doc/"
+        const linksColab = `
+        <div class="colab-dropdown">
+            <img alt="Open In Colab" src="https://colab.research.google.com/assets/colab-badge.svg">
+            <div class="colab-dropdown-content">
+                <button onclick=" window.open('${baseURL}${pageName}.ipynb')">Mixed</button>
+                <button onclick=" window.open('${baseURL}pytorch/${pageName}.ipynb')">PyTorch</button>
+                <button onclick=" window.open('${baseURL}tensorflow/${pageName}.ipynb')">TensorFlow</button>
+            </div>
+        </div>`
+        const leftMenu = document.querySelector(".wy-breadcrumbs-aside")
+        leftMenu.innerHTML = linksColab + '\n' + leftMenu.innerHTML
+    }
+}
+
 function addVersionControl() {
     // To grab the version currently in view, we parse the url
     const parts = location.toString().split('/');
     let versionIndex = parts.length - 2;
     // Index page may not have a last part with filename.html so we need to go up
-    if (parts[parts.length - 1] != "" && ! parts[parts.length - 1].match(/\.html$/)) {
+    if (parts[parts.length - 1] != "" && ! parts[parts.length - 1].match(/\.html$|^search.html?/)) {
         versionIndex = parts.length - 1;
     }
     // Main classes and models are nested so we need to go deeper
@@ -148,6 +181,7 @@ function addHfMenu() {
     <div class="menu">
         <a href="/welcome">🔥 Sign in</a>
         <a href="/models">🚀 Models</a>
+        <a href="http://discuss.huggingface.co">💬 Forum</a>
     </div>
     `;
     document.body.insertAdjacentHTML('afterbegin', div);
@@ -253,6 +287,7 @@ function onLoad() {
     addGithubButton();
     parseGithubButtons();
     addHfMenu();
+    addColabLink();
     platformToggle();
 }
 

docs/source/benchmarks.rst

@@ -40,12 +40,12 @@ There are many more parameters that can be configured via the benchmark argument
 ``src/transformers/benchmark/benchmark_args_utils.py``, ``src/transformers/benchmark/benchmark_args.py`` (for PyTorch) and ``src/transformers/benchmark/benchmark_args_tf.py`` (for Tensorflow). 
 Alternatively, running the following shell commands from root will print out a descriptive list of all configurable parameters for PyTorch and Tensorflow respectively.
 
-.. code-block::
+.. code-block:: bash
 
-    >>> ## PYTORCH CODE
+    ## PYTORCH CODE
     python examples/benchmarking/run_benchmark.py --help
 
-    >>> ## TENSORFLOW CODE
+    ## TENSORFLOW CODE
     python examples/benchmarking/run_benchmark_tf.py --help
 
 

docs/source/conf.py

@@ -26,7 +26,7 @@ author = u'huggingface'
 # The short X.Y version
 version = u''
 # The full version, including alpha/beta/rc tags
-release = u'3.0.0'
+release = u'3.1.0'
 
 
 # -- General configuration ---------------------------------------------------
@@ -76,7 +76,8 @@ exclude_patterns = [u'_build', 'Thumbs.db', '.DS_Store']
 pygments_style = None
 
 # Remove the prompt when copying examples
-copybutton_prompt_text = ">>> "
+copybutton_prompt_text = r">>> |\.\.\. "
+copybutton_prompt_is_regexp = True
 
 # -- Options for HTML output -------------------------------------------------
 

docs/source/custom_datasets.rst

@@ -0,0 +1,715 @@
+Fine-tuning with custom datasets
+================================
+
+.. note::
+
+    The datasets used in this tutorial are available and can be more easily accessed using the
+    `🤗 NLP library <https://github.com/huggingface/nlp>`_. We do not use this library to access the datasets here
+    since this tutorial meant to illustrate how to work with your own data. A brief of introduction can be found
+    at the end of the tutorial in the section ":ref:`nlplib`".
+
+This tutorial will take you through several examples of using 🤗 Transformers models with your own datasets. The
+guide shows one of many valid workflows for using these models and is meant to be illustrative rather than
+definitive. We show examples of reading in several data formats, preprocessing the data for several types of tasks,
+and then preparing the data into PyTorch/TensorFlow ``Dataset`` objects which can easily be used either with
+:class:`~transformers.Trainer`/:class:`~transformers.TFTrainer` or with native PyTorch/TensorFlow.
+
+We include several examples, each of which demonstrates a different type of common downstream task:
+
+  - :ref:`seq_imdb`
+  - :ref:`tok_ner`
+  - :ref:`qa_squad`
+  - :ref:`resources`
+
+.. _seq_imdb:
+
+Sequence Classification with IMDb Reviews
+-----------------------------------------
+
+.. note::
+
+    This dataset can be explored in the Hugging Face model hub (`IMDb <https://huggingface.co/datasets/imdb>`_), and can
+    be alternatively downloaded with the 🤗 NLP library with ``load_dataset("imdb")``.
+
+In this example, we'll show how to download, tokenize, and train a model on the IMDb reviews dataset. This task
+takes the text of a review and requires the model to predict whether the sentiment of the review is positive or
+negative. Let's start by downloading the dataset from the
+`Large Movie Review Dataset <http://ai.stanford.edu/~amaas/data/sentiment/>`_ webpage.
+
+.. code-block:: bash
+
+    wget http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz
+    tar -xf aclImdb_v1.tar.gz
+
+This data is organized into ``pos`` and ``neg`` folders with one text file per example. Let's write a function that can
+read this in.
+
+.. code-block:: python
+
+    from pathlib import Path
+
+    def read_imdb_split(split_dir):
+        split_dir = Path(split_dir)
+        texts = []
+        labels = []
+        for label_dir in ["pos", "neg"]:
+            for text_file in (split_dir/label_dir).iterdir():
+                texts.append(text_file.read_text())
+                labels.append(0 if label_dir is "neg" else 1)
+
+        return texts, labels
+
+    train_texts, train_labels = read_imdb_split('aclImdb/train')
+    test_texts, test_labels = read_imdb_split('aclImdb/test')
+
+We now have a train and test dataset, but let's also also create a validation set which we can use for for
+evaluation and tuning without training our test set results. Sklearn has a convenient utility for creating such
+splits:
+
+.. code-block:: python
+
+    from sklearn.model_selection import train_test_split
+    train_texts, val_texts, train_labels, val_labels = train_test_split(train_texts, train_labels, test_size=.2)
+
+Alright, we've read in our dataset. Now let's tackle tokenization. We'll eventually train a classifier using
+pre-trained DistilBert, so let's use the DistilBert tokenizer.
+
+.. code-block:: python
+
+    from transformers import DistilBertTokenizerFast
+    tokenizer = DistilBertTokenizerFast.from_pretrained('distilbert-base-uncased')
+
+Now we can simply pass our texts to the tokenizer. We'll pass ``truncation=True`` and ``padding=True``, which will
+ensure that all of our sequences are padded to the same length and are truncated to be no longer model's maximum
+input length. This will allow us to feed batches of sequences into the model at the same time.
+
+.. code-block:: python
+
+    train_encodings = tokenizer(train_texts, truncation=True, padding=True)
+    val_encodings = tokenizer(val_texts, truncation=True, padding=True)
+    test_encodings = tokenizer(test_texts, truncation=True, padding=True)
+
+Now, let's turn our labels and encodings into a Dataset object. In PyTorch, this is done by subclassing a
+``torch.utils.data.Dataset`` object and implementing ``__len__`` and ``__getitem__``. In TensorFlow, we pass our input encodings and
+labels to the ``from_tensor_slices`` constructor method. We put the data in this format so that the data can be
+easily batched such that each key in the batch encoding corresponds to a named parameter of the
+:meth:`~transformers.DistilBertForSequenceClassification.forward` method of the model we will train.
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    import torch
+
+    class IMDbDataset(torch.utils.data.Dataset):
+        def __init__(self, encodings, labels):
+            self.encodings = encodings
+            self.labels = labels
+
+        def __getitem__(self, idx):
+            item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
+            item['labels'] = torch.tensor(self.labels[idx])
+            return item
+
+        def __len__(self):
+            return len(self.labels)
+
+    train_dataset = IMDbDataset(train_encodings, train_labels)
+    val_dataset = IMDbDataset(val_encodings, val_labels)
+    test_dataset = IMDbDataset(test_encodings, test_labels)
+    ## TENSORFLOW CODE
+    import tensorflow as tf
+
+    train_dataset = tf.data.Dataset.from_tensor_slices((
+        dict(train_encodings),
+        train_labels
+    ))
+    val_dataset = tf.data.Dataset.from_tensor_slices((
+        dict(val_encodings),
+        val_labels
+    ))
+    test_dataset = tf.data.Dataset.from_tensor_slices((
+        dict(test_encodings),
+        test_labels
+    ))
+
+Now that our datasets our ready, we can fine-tune a model either with the 🤗
+:class:`~transformers.Trainer`/:class:`~transformers.TFTrainer` or with native PyTorch/TensorFlow. See
+:doc:`training <training>`.
+
+.. _ft_trainer:
+
+Fine-tuning with Trainer
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+The steps above prepared the datasets in the way that the trainer is expected. Now all we need to do is create a
+model to fine-tune, define the :class:`~transformers.TrainingArguments`/:class:`~transformers.TFTrainingArguments`
+and instantiate a :class:`~transformers.Trainer`/:class:`~transformers.TFTrainer`.
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    from transformers import DistilBertForSequenceClassification, Trainer, TrainingArguments
+
+    training_args = TrainingArguments(
+        output_dir='./results',          # output directory
+        num_train_epochs=3,              # total number of training epochs
+        per_device_train_batch_size=16,  # batch size per device during training
+        per_device_eval_batch_size=64,   # batch size for evaluation
+        warmup_steps=500,                # number of warmup steps for learning rate scheduler
+        weight_decay=0.01,               # strength of weight decay
+        logging_dir='./logs',            # directory for storing logs
+        logging_steps=10,
+    )
+
+    model = DistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased")
+
+    trainer = Trainer(
+        model=model,                         # the instantiated 🤗 Transformers model to be trained
+        args=training_args,                  # training arguments, defined above
+        train_dataset=train_dataset,         # training dataset
+        eval_dataset=val_dataset             # evaluation dataset
+    )
+
+    trainer.train()
+    ## TENSORFLOW CODE
+    from transformers import TFDistilBertForSequenceClassification, TFTrainer, TFTrainingArguments
+
+    training_args = TFTrainingArguments(
+        output_dir='./results',          # output directory
+        num_train_epochs=3,              # total number of training epochs
+        per_device_train_batch_size=16,  # batch size per device during training
+        per_device_eval_batch_size=64,   # batch size for evaluation
+        warmup_steps=500,                # number of warmup steps for learning rate scheduler
+        weight_decay=0.01,               # strength of weight decay
+        logging_dir='./logs',            # directory for storing logs
+        logging_steps=10,
+    )
+
+    with training_args.strategy.scope():
+        model = TFDistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased")
+
+    trainer = TFTrainer(
+        model=model,                         # the instantiated 🤗 Transformers model to be trained
+        args=training_args,                  # training arguments, defined above
+        train_dataset=train_dataset,         # training dataset
+        eval_dataset=val_dataset             # evaluation dataset
+    )
+
+    trainer.train()
+
+.. _ft_native:
+
+Fine-tuning with native PyTorch/TensorFlow
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+We can also train use native PyTorch or TensorFlow:
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    from torch.utils.data import DataLoader
+    from transformers import DistilBertForSequenceClassification, AdamW
+
+    device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
+
+    model = DistilBertForSequenceClassification.from_pretrained('distilbert-base-uncased')
+    model.to(device)
+    model.train()
+
+    train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)
+
+    optim = AdamW(model.parameters(), lr=5e-5)
+
+    for epoch in range(3):
+        for batch in train_loader:
+            optim.zero_grad()
+            input_ids = batch['input_ids'].to(device)
+            attention_mask = batch['attention_mask'].to(device)
+            labels = batch['labels'].to(device)
+            outputs = model(input_ids, attention_mask=attention_mask, labels=labels)
+            loss = outputs[0]
+            loss.backward()
+            optim.step()
+
+    model.eval()
+    ## TENSORFLOW CODE
+    from transformers import TFDistilBertForSequenceClassification
+
+    model = TFDistilBertForSequenceClassification.from_pretrained('distilbert-base-uncased')
+
+    optimizer = tf.keras.optimizers.Adam(learning_rate=5e-5)
+    model.compile(optimizer=optimizer, loss=model.compute_loss) # can also use any keras loss fn
+    model.fit(train_dataset.shuffle(1000).batch(16), epochs=3, batch_size=16)
+
+.. _tok_ner:
+
+Token Classification with W-NUT Emerging Entities
+-------------------------------------------------
+
+.. note::
+
+    This dataset can be explored in the Hugging Face model hub (`WNUT-17 <https://huggingface.co/datasets/wnut_17>`_), and can
+    be alternatively downloaded with the 🤗 NLP library with ``load_dataset("wnut_17")``.
+
+Next we will look at token classification. Rather than classifying an entire sequence, this task classifies token by
+token. We'll demonstrate how to do this with 
+`Named Entity Recognition <http://nlpprogress.com/english/named_entity_recognition.html>`_, which involves
+identifying tokens which correspond to a predefined set of "entities". Specifically, we'll use the
+`W-NUT Emerging and Rare entities <http://noisy-text.github.io/2017/emerging-rare-entities.html>`_ corpus. The data
+is given as a collection of pre-tokenized documents where each token is assigned a tag.
+
+Let's start by downloading the data.
+
+.. code-block:: bash
+
+    wget http://noisy-text.github.io/2017/files/wnut17train.conll
+
+In this case, we'll just download the train set, which is a single text file. Each line of the file contains either
+(1) a word and tag separated by a tab, or (2) a blank line indicating the end of a document. Let's write a
+function to read this in. We'll take in the file path and return ``token_docs`` which is a list of lists of token
+strings, and ``token_tags`` which is a list of lists of tag strings.
+
+.. code-block:: python
+
+    from pathlib import Path
+    import re
+
+    def read_wnut(file_path):
+        file_path = Path(file_path)
+
+        raw_text = file_path.read_text().strip()
+        raw_docs = re.split(r'\n\t?\n', raw_text)
+        token_docs = []
+        tag_docs = []
+        for doc in raw_docs:
+            tokens = []
+            tags = []
+            for line in doc.split('\n'):
+                token, tag = line.split('\t')
+                tokens.append(token)
+                tags.append(tag)
+            token_docs.append(tokens)
+            tag_docs.append(tags)
+        
+        return token_docs, tag_docs
+    
+    texts, tags = read_wnut('wnut17train.conll')
+    
+Just to see what this data looks like, let's take a look at a segment of the first document.
+
+.. code-block:: python
+
+    >>> print(texts[0][10:17], tags[0][10:17], sep='\n')
+    ['for', 'two', 'weeks', '.', 'Empire', 'State', 'Building']
+    ['O', 'O', 'O', 'O', 'B-location', 'I-location', 'I-location']
+
+``location`` is an entity type, ``B-`` indicates the beginning of an entity, and ``I-`` indicates consecutive positions of
+the same entity ("Empire State Building" is considered one entity). ``O`` indicates the token does not correspond to
+any entity.
+
+Now that we've read the data in, let's create a train/validation split:
+
+.. code-block:: python
+
+    from sklearn.model_selection import train_test_split
+    train_texts, val_texts, train_tags, val_tags = train_test_split(texts, tags, test_size=.2)
+
+Next, let's create encodings for our tokens and tags. For the tags, we can start by just create a simple mapping
+which we'll use in a moment:
+
+.. code-block:: python
+
+    unique_tags = set(tag for doc in tags for tag in doc)
+    tag2id = {tag: id for id, tag in enumerate(unique_tags)}
+    id2tag = {id: tag for tag, id in tag2id.items()}
+
+To encode the tokens, we'll use a pre-trained DistilBert tokenizer. We can tell the tokenizer that we're dealing
+with ready-split tokens rather than full sentence strings by passing ``is_pretokenized=True``. We'll also pass
+``padding=True`` and ``truncation=True`` to pad the sequences to be the same length. Lastly, we can tell the model
+to return information about the tokens which are split by the wordpiece tokenization process, which we will need in
+a moment.
+
+.. code-block:: python
+
+    from transformers import DistilBertTokenizerFast
+    tokenizer = DistilBertTokenizerFast.from_pretrained('distilbert-base-cased')
+    train_encodings = tokenizer(train_texts, is_pretokenized=True, return_offsets_mapping=True, padding=True, truncation=True)
+    val_encodings = tokenizer(val_texts, is_pretokenized=True, return_offsets_mapping=True, padding=True, truncation=True)
+
+Great, so now our tokens are nicely encoded in the format that they need to be in to feed them into our DistilBert
+model below.
+
+Now we arrive at a common obstacle with using pre-trained models for token-level classification: many of the tokens
+in the W-NUT corpus are not in DistilBert's vocabulary. Bert and many models like it use a method called WordPiece
+Tokenization, meaning that single words are split into multiple tokens such that each token is likely to be in
+the vocabulary. For example, DistilBert's tokenizer would split the Twitter handle ``@huggingface`` into the tokens
+``['@', 'hugging', '##face']``. This is a problem for us because we have exactly one tag per token. If the tokenizer
+splits a token into multiple sub-tokens, then we will end up with a mismatch between our tokens and our labels.
+
+One way to handle this is to only train on the tag labels for the first subtoken of a split token. We can do this in
+🤗 Transformers by setting the labels we wish to ignore to ``-100``. In the example above, if the label for
+``@HuggingFace`` is ``3`` (indexing ``B-corporation``), we would set the labels of ``['@', 'hugging', '##face']`` to
+``[3, -100, -100]``.
+
+Let's write a function to do this. This is where we will use the ``offset_mapping`` from the tokenizer as mentioned
+above. For each sub-token returned by the tokenizer, the offset mapping gives us a tuple indicating the sub-token's
+start position and end position relative to the original token it was split from. That means that if the first
+position in the tuple is anything other than ``0``, we will set its corresponding label to ``-100``. While we're at
+it, we can also set labels to ``-100`` if the second position of the offset mapping is ``0``, since this means it must
+be a special token like ``[PAD]`` or ``[CLS]``.
+
+.. note:: 
+
+    Due to a recently fixed bug, -1 must be used instead of -100 when using TensorFlow in 🤗 Transformers <= 3.02.
+
+.. code-block:: python
+
+    import numpy as np
+
+    def encode_tags(tags, encodings):
+        labels = [[tag2id[tag] for tag in doc] for doc in tags]
+        encoded_labels = []
+        for doc_labels, doc_offset in zip(labels, encodings.offset_mapping):
+            # create an empty array of -100
+            doc_enc_labels = np.ones(len(doc_offset),dtype=int) * -100
+            arr_offset = np.array(doc_offset)
+
+            # set labels whose first offset position is 0 and the second is not 0
+            doc_enc_labels[(arr_offset[:,0] == 0) & (arr_offset[:,1] != 0)] = doc_labels
+            encoded_labels.append(doc_enc_labels.tolist())
+
+        return encoded_labels
+    
+    train_labels = encode_tags(train_tags, train_encodings)
+    val_labels = encode_tags(val_tags, val_encodings)
+
+The hard part is now done. Just as in the sequence classification example above, we can create a dataset object:
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    import torch
+
+    class WNUTDataset(torch.utils.data.Dataset):
+        def __init__(self, encodings, labels):
+            self.encodings = encodings
+            self.labels = labels
+
+        def __getitem__(self, idx):
+            item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
+            item['labels'] = torch.tensor(self.labels[idx])
+            return item
+
+        def __len__(self):
+            return len(self.labels)
+
+    train_encodings.pop("offset_mapping") # we don't want to pass this to the model
+    val_encodings.pop("offset_mapping")
+    train_dataset = WNUTDataset(train_encodings, train_labels)
+    val_dataset = WNUTDataset(val_encodings, val_labels)
+    ## TENSORFLOW CODE
+    import tensorflow as tf
+
+    train_encodings.pop("offset_mapping") # we don't want to pass this to the model
+    val_encodings.pop("offset_mapping")
+
+    train_dataset = tf.data.Dataset.from_tensor_slices((
+        dict(train_encodings),
+        train_labels
+    ))
+    val_dataset = tf.data.Dataset.from_tensor_slices((
+        dict(val_encodings),
+        val_labels
+    ))
+
+Now load in a token classification model and specify the number of labels:
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    from transformers import DistilBertForTokenClassification
+    model = DistilBertForTokenClassification.from_pretrained('distilbert-base-cased', num_labels=len(unique_tags))
+    ## TENSORFLOW CODE
+    from transformers import TFDistilBertForTokenClassification
+    model = TFDistilBertForTokenClassification.from_pretrained('distilbert-base-cased', num_labels=len(unique_tags))
+
+The data and model are both ready to go. You can train the model either with
+:class:`~transformers.Trainer`/:class:`~transformers.TFTrainer` or with native PyTorch/TensorFlow, exactly as in the
+sequence classification example above.
+
+  - :ref:`ft_trainer`
+  - :ref:`ft_native`
+
+.. _qa_squad:
+
+Question Answering with SQuAD 2.0
+---------------------------------
+
+.. note::
+
+    This dataset can be explored in the Hugging Face model hub (`SQuAD V2 <https://huggingface.co/datasets/squad_v2>`_), and can
+    be alternatively downloaded with the 🤗 NLP library with ``load_dataset("squad_v2")``.
+
+Question answering comes in many forms. In this example, we'll look at the particular type of extractive QA that
+involves answering a question about a passage by highlighting the segment of the passage that answers the question.
+This involves fine-tuning a model which predicts a start position and an end position in the passage. We will use the
+`Stanford Question Answering Dataset (SQuAD) 2.0 <https://rajpurkar.github.io/SQuAD-explorer/>`_.
+
+We will start by downloading the data:
+
+.. code-block:: bash
+
+    mkdir squad
+    wget https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v2.0.json -O squad/train-v2.0.json
+    wget https://rajpurkar.github.io/SQuAD-explorer/dataset/dev-v2.0.json -O squad/dev-v2.0.json
+
+Each split is in a structured json file with a number of questions and answers for each passage (or context). We'll
+take this apart into parallel lists of contexts, questions, and answers (note that the contexts here are repeated
+since there are multiple questions per context):
+
+.. code-block:: python
+
+    import json
+    from pathlib import Path
+
+    def read_squad(path):
+        path = Path(path)
+        with open(path, 'rb') as f:
+            squad_dict = json.load(f)
+
+        contexts = []
+        questions = []
+        answers = []
+        for group in squad_dict['data']:
+            for passage in group['paragraphs']:
+                context = passage['context']
+                for qa in passage['qas']:
+                    question = qa['question']
+                    for answer in qa['answers']:
+                        contexts.append(context)
+                        questions.append(question)
+                        answers.append(answer)
+
+        return contexts, questions, answers
+    
+    train_contexts, train_questions, train_answers = read_squad('squad/train-v2.0.json')
+    val_contexts, val_questions, val_answers = read_squad('squad/dev-v2.0.json')
+
+The contexts and questions are just strings. The answers are dicts containing the subsequence of the passage with
+the correct answer as well as an integer indicating the character at which the answer begins. In order to train a
+model on this data we need (1) the tokenized context/question pairs, and (2) integers indicating at which *token*
+positions the answer begins and ends.
+
+First, let's get the *character* position at which the answer ends in the passage (we are given the starting
+position). Sometimes SQuAD answers are off by one or two characters, so we will also adjust for that.
+
+.. code-block:: python
+
+    def add_end_idx(answers, contexts):
+        for answer, context in zip(answers, contexts):
+            gold_text = answer['text']
+            start_idx = answer['answer_start']
+            end_idx = start_idx + len(gold_text)
+            
+            # sometimes squad answers are off by a character or two – fix this
+            if context[start_idx:end_idx] == gold_text:
+                answer['answer_end'] = end_idx
+            elif context[start_idx-1:end_idx-1] == gold_text:
+                answer['answer_start'] = start_idx - 1
+                answer['answer_end'] = end_idx - 1     # When the gold label is off by one character
+            elif context[start_idx-2:end_idx-2] == gold_text:
+                answer['answer_start'] = start_idx - 2
+                answer['answer_end'] = end_idx - 2     # When the gold label is off by two characters
+
+    add_end_idx(train_answers, train_contexts)
+    add_end_idx(val_answers, val_contexts)
+
+Now ``train_answers`` and ``val_answers`` include the character end positions and the corrected start positions.
+Next, let's tokenize our context/question pairs. 🤗 Tokenizers can accept parallel lists of sequences and encode
+them together as sequence pairs.
+
+.. code-block:: python
+
+    from transformers import DistilBertTokenizerFast
+    tokenizer = DistilBertTokenizerFast.from_pretrained('distilbert-base-uncased')
+
+    train_encodings = tokenizer(train_contexts, train_questions, truncation=True, padding=True)
+    val_encodings = tokenizer(val_contexts, val_questions, truncation=True, padding=True)
+
+Next we need to convert our character start/end positions to token start/end positions. When using 🤗 Fast
+Tokenizers, we can use the built in :func:`~transformers.BatchEncoding.char_to_token` method.
+
+.. code-block:: python
+
+    def add_token_positions(encodings, answers):
+        start_positions = []
+        end_positions = []
+        for i in range(len(answers)):
+            start_positions.append(encodings.char_to_token(i, answers[i]['answer_start']))
+            end_positions.append(encodings.char_to_token(i, answers[i]['answer_end'] - 1))
+            # if None, the answer passage has been truncated
+            if start_positions[-1] is None:
+                start_positions[-1] = tokenizer.model_max_length
+            if end_positions[-1] is None:
+                end_positions[-1] = tokenizer.model_max_length
+        encodings.update({'start_positions': start_positions, 'end_positions': end_positions})
+
+    add_token_positions(train_encodings, train_answers)
+    add_token_positions(val_encodings, val_answers)
+
+Our data is ready. Let's just put it in a PyTorch/TensorFlow dataset so that we can easily use it for
+training. In PyTorch, we define a custom ``Dataset`` class. In TensorFlow, we pass a tuple of
+``(inputs_dict, labels_dict)`` to the ``from_tensor_slices`` method.
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    import torch
+
+    class SquadDataset(torch.utils.data.Dataset):
+        def __init__(self, encodings):
+            self.encodings = encodings
+
+        def __getitem__(self, idx):
+            return {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
+
+        def __len__(self):
+            return len(self.encodings.input_ids)
+        
+    train_dataset = SquadDataset(train_encodings)
+    val_dataset = SquadDataset(val_encodings)
+    ## TENSORFLOW CODE
+    import tensorflow as tf
+
+    train_dataset = tf.data.Dataset.from_tensor_slices((
+        {key: train_encodings[key] for key in ['input_ids', 'attention_mask']},
+        {key: train_encodings[key] for key in ['start_positions', 'end_positions']}
+    ))
+    val_dataset = tf.data.Dataset.from_tensor_slices((
+        {key: val_encodings[key] for key in ['input_ids', 'attention_mask']},
+        {key: val_encodings[key] for key in ['start_positions', 'end_positions']}
+    ))
+
+Now we can use a DistilBert model with a QA head for training:
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    from transformers import DistilBertForQuestionAnswering
+    model = DistilBertForQuestionAnswering.from_pretrained("distilbert-base-uncased")
+    ## TENSORFLOW CODE
+    from transformers import TFDistilBertForQuestionAnswering
+    model = TFDistilBertForQuestionAnswering.from_pretrained("distilbert-base-uncased")
+
+
+The data and model are both ready to go. You can train the model with
+:class:`~transformers.Trainer`/:class:`~transformers.TFTrainer` exactly as in the sequence classification example
+above. If using native PyTorch, replace ``labels`` with ``start_positions`` and ``end_positions`` in the training
+example. If using Keras's ``fit``, we need to make a minor modification to handle this example since it involves
+multiple model outputs.
+
+  - :ref:`ft_trainer`
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    from torch.utils.data import DataLoader
+    from transformers import AdamW
+
+    device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
+
+    model.to(device)
+    model.train()
+
+    train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)
+
+    optim = AdamW(model.parameters(), lr=5e-5)
+
+    for epoch in range(3):
+        for batch in train_loader:
+            optim.zero_grad()
+            input_ids = batch['input_ids'].to(device)
+            attention_mask = batch['attention_mask'].to(device)
+            start_positions = batch['start_positions'].to(device)
+            end_positions = batch['end_positions'].to(device)
+            outputs = model(input_ids, attention_mask=attention_mask, start_positions=start_positions, end_positions=end_positions)
+            loss = outputs[0]
+            loss.backward()
+            optim.step()
+
+    model.eval()
+    ## TENSORFLOW CODE
+    # Keras will expect a tuple when dealing with labels
+    train_dataset = train_dataset.map(lambda x, y: (x, (y['start_positions'], y['end_positions'])))
+
+    # Keras will assign a separate loss for each output and add them together. So we'll just use the standard CE loss
+    # instead of using the built-in model.compute_loss, which expects a dict of outputs and averages the two terms.
+    # Note that this means the loss will be 2x of when using TFTrainer since we're adding instead of averaging them.
+    loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
+    model.distilbert.return_dict = False # if using 🤗 Transformers >3.02, make sure outputs are tuples
+
+    optimizer = tf.keras.optimizers.Adam(learning_rate=5e-5)
+    model.compile(optimizer=optimizer, loss=loss) # can also use any keras loss fn
+    model.fit(train_dataset.shuffle(1000).batch(16), epochs=3, batch_size=16)
+
+.. _resources:
+
+Additional Resources
+--------------------
+
+  - `How to train a new language model from scratch using Transformers and Tokenizers
+    <https://huggingface.co/blog/how-to-train>`_. Blog post showing the steps to load in Esperanto data and train a
+    masked language model from scratch.
+  - :doc:`Preprocessing <preprocessing>`. Docs page on data preprocessing.
+  - :doc:`Training <training>`. Docs page on training and fine-tuning.
+
+.. _nlplib:
+
+Using the 🤗 NLP Datasets & Metrics library
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+This tutorial demonstrates how to read in datasets from various raw text formats and prepare them for training with
+🤗 Transformers so that you can do the same thing with your own custom datasets. However, we recommend users use the
+`🤗 NLP library <https://github.com/huggingface/nlp>`_ for working with the 150+ datasets included in the
+`hub <https://huggingface.co/datasets>`_, including the three datasets used in this tutorial. As a very brief overview,
+we will show how to use the NLP library to download and prepare the IMDb dataset from the first example,
+:ref:`seq_imdb`.
+
+Start by downloading the dataset:
+
+.. code-block:: python
+
+    from nlp import load_dataset
+    train = load_dataset("imdb", split="train")
+
+Each dataset has multiple columns corresponding to different features. Let's see what our columns are.
+
+.. code-block:: python
+
+    >>> print(train.column_names)
+    ['label', 'text']
+
+Great. Now let's tokenize the text. We can do this using the ``map`` method. We'll also rename the ``label`` column
+to ``labels`` to match the model's input arguments.
+
+.. code-block:: python
+
+    train = train.map(lambda batch: tokenizer(batch["text"], truncation=True, padding=True), batched=True)
+    train.rename_column_("label", "labels")
+
+Lastly, we can use the ``set_format`` method to determine which columns and in what data format we want to access
+dataset elements.
+
+.. code-block:: python
+
+    ## PYTORCH CODE
+    >>> train.set_format("torch", columns=["input_ids", "attention_mask", "labels"])
+    >>> {key: val.shape for key, val in train[0].items()})
+    {'labels': torch.Size([]), 'input_ids': torch.Size([512]), 'attention_mask': torch.Size([512])}
+    ## TENSORFLOW CODE
+    >>> train.set_format("tensorflow", columns=["input_ids", "attention_mask", "labels"])
+    >>> {key: val.shape for key, val in train[0].items()})
+    {'labels': TensorShape([]), 'input_ids': TensorShape([512]), 'attention_mask': TensorShape([512])}
+
+We now have a fully-prepared dataset. Check out `the 🤗 NLP docs <https://huggingface.co/nlp/processing.html>`_ for
+a more thorough introduction.

docs/source/glossary.rst

@@ -7,18 +7,18 @@ General terms
 - autoencoding models: see MLM
 - autoregressive models: see CLM
 - CLM: causal language modeling, a pretraining task where the model reads the texts in order and has to predict the
-  next word. It's usually done by reading the whole sentence but using a mask inside the model to hide the future 
+  next word. It's usually done by reading the whole sentence but using a mask inside the model to hide the future
   tokens at a certain timestep.
 - MLM: masked language modeling, a pretraining task where the model sees a corrupted version of the texts, usually done
   by masking some tokens randomly, and has to predict the original text.
-- multimodal: a task taht combines texts with another kind of inputs (for instance images).
+- multimodal: a task that combines texts with another kind of inputs (for instance images).
 - NLG: natural language generation, all tasks related to generating text ( for instance talk with transformers,
   translation)
 - NLP: natural language processing, a generic way to say "deal with texts".
 - NLU: natural language understanding, all tasks related to understanding what is in a text (for instance classifying
   the whole text, individual words)
 - pretrained model: a model that has been pretrained on some data (for instance all of Wikipedia). Pretraining methods
-  involve a self-supervised objective, which can be reading the text and trying to predict the next word (see CLM) or 
+  involve a self-supervised objective, which can be reading the text and trying to predict the next word (see CLM) or
   masking some words and trying to predict them (see MLM).
 - RNN: recurrent neural network, a type of model that uses a loop over a layer to process texts.
 - seq2seq or sequence-to-sequence: models that generate a new sequence from an input, like translation models, or
@@ -57,7 +57,7 @@ The tokenizer takes care of splitting the sequence into tokens available in the
     >>> tokenized_sequence = tokenizer.tokenize(sequence)
 
 The tokens are either words or subwords. Here for instance, "VRAM" wasn't in the model vocabulary, so it's been split
-in "V", "RA" and "M". To indicate those tokens are not separate words but parts of the same word, a double-dash is
+in "V", "RA" and "M". To indicate those tokens are not separate words but parts of the same word, a double-hash prefix is
 added for "RA" and "M":
 
 ::
@@ -71,24 +71,27 @@ the sentence to the tokenizer, which leverages the Rust implementation of
 
 ::
 
-    >>> encoded_sequence = tokenizer(sequence)["input_ids"]
+    >>> inputs = tokenizer(sequence)
 
 The tokenizer returns a dictionary with all the arguments necessary for its corresponding model to work properly. The
 token indices are under the key "input_ids":
 
 ::
 
+    >>> encoded_sequence = inputs["input_ids"]
     >>> print(encoded_sequence)
     [101, 138, 18696, 155, 1942, 3190, 1144, 1572, 13745, 1104, 159, 9664, 2107, 102]
 
-Note that the tokenizer automatically adds "special tokens" (if the associated model rely on them) which are special
-IDs the model sometimes uses. If we decode the previous sequence of ids,
+Note that the tokenizer automatically adds "special tokens" (if the associated model relies on them) which are special
+IDs the model sometimes uses.
+
+If we decode the previous sequence of ids,
 
 ::
 
     >>> decoded_sequence = tokenizer.decode(encoded_sequence)
 
-we will see 
+we will see
 
 ::
 
@@ -144,7 +147,7 @@ We can see that 0s have been added on the right of the first sentence to make it
 
 This can then be converted into a tensor in PyTorch or TensorFlow. The attention mask is a binary tensor indicating
 the position of the padded indices so that the model does not attend to them. For the
-:class:`~transformers.BertTokenizer`, :obj:`1` indicate a value that should be attended to while :obj:`0` indicate
+:class:`~transformers.BertTokenizer`, :obj:`1` indicates a value that should be attended to, while :obj:`0` indicates
 a padded value. This attention mask is in the dictionary returned by the tokenizer under the key "attention_mask":
 
 ::
@@ -158,15 +161,15 @@ Token Type IDs
 ~~~~~~~~~~~~~~
 
 Some models' purpose is to do sequence classification or question answering. These require two different sequences to
-be encoded in the same input IDs. They are usually separated by special tokens, such as the classifier and separator
+be joined in a single "input_ids" entry, which usually is performed with the help of special tokens, such as the classifier (``[CLS]``) and separator (``[SEP]``)
 tokens. For example, the BERT model builds its two sequence input as such:
 
 ::
 
    >>> # [CLS] SEQUENCE_A [SEP] SEQUENCE_B [SEP]
 
-We can use our tokenizer to automatically generate such a sentence by passing the two sequences as two arguments (and
-not a list like before) like this:
+We can use our tokenizer to automatically generate such a sentence by passing the two sequences to ``tokenizer`` as two arguments (and
+not a list, like before) like this:
 
 ::
 
@@ -185,31 +188,31 @@ which will return:
     >>> print(decoded)
     [CLS] HuggingFace is based in NYC [SEP] Where is HuggingFace based? [SEP]
 
-This is enough for some models to understand where one sequence ends and where another begins. However, other models
-such as BERT have an additional mechanism, which are the token type IDs (also called segment IDs). They are a binary
-mask identifying the different sequences in the model.
+This is enough for some models to understand where one sequence ends and where another begins. However, other models,
+such as BERT, also deploy token type IDs (also called segment IDs). They are represented as a binary
+mask identifying the two types of sequence in the model.
 
-The tokenizer returns in the dictionary under the key "token_type_ids":
+The tokenizer returns this mask as the "token_type_ids" entry:
 
 ::
 
     >>> encoded_dict['token_type_ids']
     [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1]
 
-The first sequence, the "context" used for the question, has all its tokens represented by :obj:`0`, whereas the
-question has all its tokens represented by :obj:`1`. Some models, like :class:`~transformers.XLNetModel` use an
-additional token represented by a :obj:`2`.
+The first sequence, the "context" used for the question, has all its tokens represented by a :obj:`0`, whereas the
+second sequence, corresponding to the "question", has all its tokens represented by a :obj:`1`.
+
+Some models, like :class:`~transformers.XLNetModel` use an additional token represented by a :obj:`2`.
 
 .. _position-ids:
 
 Position IDs
 ~~~~~~~~~~~~
 
-The position IDs are used by the model to identify which token is at which position. Contrary to RNNs that have the
-position of each token embedded within them, transformers are unaware of the position of each token. The position
-IDs are created for this purpose.
+Contrary to RNNs that have the position of each token embedded within them,
+transformers are unaware of the position of each token. Therefore, the position IDs (``position_ids``) are used by the model to identify each token's position in the list of tokens.
 
-They are an optional parameter. If no position IDs are passed to the model, they are automatically created as absolute
+They are an optional parameter. If no ``position_ids`` is passed to the model, the IDs are automatically created as absolute
 positional embeddings.
 
 Absolute positional embeddings are selected in the range ``[0, config.max_position_embeddings - 1]``. Some models
@@ -220,15 +223,15 @@ use other types of positional embeddings, such as sinusoidal position embeddings
 Feed Forward Chunking
 ~~~~~~~~~~~~~~~~~~~~~
 
-In transformers two feed forward layers usually follows the self attention layer in each residual attention block.
+In each residual attention block in transformers the self-attention layer is usually followed by 2 feed forward layers.
 The intermediate embedding size of the feed forward layers is often bigger than the hidden size of the model (e.g.,
-for ``bert-base-uncased``). 
+for ``bert-base-uncased``).
 
 For an input of size ``[batch_size, sequence_length]``, the memory required to store the intermediate feed forward
 embeddings ``[batch_size, sequence_length, config.intermediate_size]`` can account for a large fraction of the memory
 use. The authors of `Reformer: The Efficient Transformer <https://arxiv.org/abs/2001.04451>`_ noticed that since the
 computation is independent of the ``sequence_length`` dimension, it is mathematically equivalent to compute the output
-embeddings of both feed forward layers ``[batch_size, config.hidden_size]_0, ..., [batch_size, config.hidden_size]_n`` 
+embeddings of both feed forward layers ``[batch_size, config.hidden_size]_0, ..., [batch_size, config.hidden_size]_n``
 individually and concat them afterward to ``[batch_size, sequence_length, config.hidden_size]`` with
 ``n = sequence_length``, which trades increased computation time against reduced memory use, but yields a
 mathematically **equivalent** result.

docs/source/index.rst

@@ -121,7 +121,14 @@ conversion utilities for the following models:
     trained using `OPUS <http://opus.nlpl.eu/>`_ pretrained_models data by Jörg Tiedemann.
 21. `Longformer <https://github.com/allenai/longformer>`_ (from AllenAI) released with the paper `Longformer: The
     Long-Document Transformer <https://arxiv.org/abs/2004.05150>`_ by Iz Beltagy, Matthew E. Peters, and Arman Cohan.
-22. `Other community models <https://huggingface.co/models>`_, contributed by the `community
+22. `DPR <https://github.com/facebookresearch/DPR>`_ (from Facebook) released with the paper `Dense Passage Retrieval
+    for Open-Domain Question Answering <https://arxiv.org/abs/2004.04906>`_ by Vladimir Karpukhin, Barlas Oğuz, Sewon
+    Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih.
+23. `Pegasus <https://github.com/google-research/pegasus>`_ (from Google) released with the paper `PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
+    <https://arxiv.org/abs/1912.08777>`_ by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu.
+24. `MBart <https://github.com/pytorch/fairseq/tree/master/examples/mbart>`_ (from Facebook) released with the paper  `Multilingual Denoising Pre-training for Neural Machine Translation <https://arxiv.org/abs/2001.08210>`_ by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov,
+    Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer.  
+25. `Other community models <https://huggingface.co/models>`_, contributed by the `community
     <https://huggingface.co/users>`_.
 
 .. toctree::
@@ -142,6 +149,7 @@ conversion utilities for the following models:
     preprocessing
     training
     model_sharing
+    tokenizer_summary
     multilingual
 
 .. toctree::
@@ -150,17 +158,19 @@ conversion utilities for the following models:
 
     pretrained_models
     examples
+    custom_datasets
     notebooks
     converting_tensorflow_models
     migration
-    torchscript
     contributing
+    serialization
 
 .. toctree::
     :maxdepth: 2
     :caption: Research
 
     bertology
+    perplexity
     benchmarks
 
 .. toctree::
@@ -168,11 +178,14 @@ conversion utilities for the following models:
     :caption: Package Reference
 
     main_classes/configuration
+    main_classes/output
     main_classes/model
     main_classes/tokenizer
     main_classes/pipelines
+    main_classes/trainer
     main_classes/optimizer_schedules
     main_classes/processors
+    main_classes/logging
     model_doc/auto
     model_doc/encoderdecoder
     model_doc/bert
@@ -197,3 +210,9 @@ conversion utilities for the following models:
     model_doc/longformer
     model_doc/retribert
     model_doc/mobilebert
+    model_doc/dpr
+    model_doc/pegasus
+    model_doc/mbart
+    internal/modeling_utils
+    internal/tokenization_utils
+    internal/pipelines_utils

docs/source/installation.md

@@ -22,13 +22,13 @@ When TensorFlow 2.0 and/or PyTorch has been installed, 🤗 Transformers can be
 pip install transformers
 ```
 
-Alternatively, for CPU-support only, you can install 🤗 Transformers and PyTorch in one line with
+Alternatively, for CPU-support only, you can install 🤗 Transformers and PyTorch in one line with:
 
 ```bash
 pip install transformers[torch]
 ```
 
-or 🤗 Transformers and TensorFlow 2.0 in one line with
+or 🤗 Transformers and TensorFlow 2.0 in one line with:
 
 ```bash
 pip install transformers[tf-cpu]
@@ -73,8 +73,8 @@ This library provides pretrained models that will be downloaded and cached local
 folder given by the shell environment variable ``TRANSFORMERS_CACHE``. The default value for it will be the PyTorch
 cache home followed by ``/transformers/`` (even if you don't have PyTorch installed). This is (by order of priority):
 
-  * shell environment variable ``ENV_TORCH_HOME``
-  * shell environment variable ``ENV_XDG_CACHE_HOME`` + ``/torch/``
+  * shell environment variable ``TORCH_HOME``
+  * shell environment variable ``XDG_CACHE_HOME`` + ``/torch/``
   * default: ``~/.cache/torch/``
 
 So if you don't have any specific environment variable set, the cache directory will be at

docs/source/internal/modeling_utils.rst

@@ -0,0 +1,88 @@
+Custom Layers and Utilities
+---------------------------
+
+This page lists all the custom layers used by the library, as well as the utility functions it provides for modeling.
+
+Most of those are only useful if you are studying the code of the models in the library.
+
+
+``Pytorch custom modules``
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_utils.Conv1D
+
+.. autoclass:: transformers.modeling_utils.PoolerStartLogits
+    :members: forward
+
+.. autoclass:: transformers.modeling_utils.PoolerEndLogits
+    :members: forward
+
+.. autoclass:: transformers.modeling_utils.PoolerAnswerClass
+    :members: forward
+
+.. autoclass:: transformers.modeling_utils.SquadHeadOutput
+
+.. autoclass:: transformers.modeling_utils.SQuADHead
+    :members: forward
+
+.. autoclass:: transformers.modeling_utils.SequenceSummary
+    :members: forward
+
+
+``PyTorch Helper Functions``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autofunction:: transformers.apply_chunking_to_forward
+
+.. autofunction:: transformers.modeling_utils.find_pruneable_heads_and_indices
+
+.. autofunction:: transformers.modeling_utils.prune_layer
+
+.. autofunction:: transformers.modeling_utils.prune_conv1d_layer
+
+.. autofunction:: transformers.modeling_utils.prune_linear_layer
+
+``TensorFlow custom layers``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_tf_utils.TFConv1D
+
+.. autoclass:: transformers.modeling_tf_utils.TFSharedEmbeddings
+    :members: call
+
+.. autoclass:: transformers.modeling_tf_utils.TFSequenceSummary
+    :members: call
+
+
+``TensorFlow loss functions``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_tf_utils.TFCausalLanguageModelingLoss
+    :members:
+
+.. autoclass:: transformers.modeling_tf_utils.TFMaskedLanguageModelingLoss
+    :members:
+
+.. autoclass:: transformers.modeling_tf_utils.TFMultipleChoiceLoss
+    :members:
+
+.. autoclass:: transformers.modeling_tf_utils.TFQuestionAnsweringLoss
+    :members:
+
+.. autoclass:: transformers.modeling_tf_utils.TFSequenceClassificationLoss
+    :members:
+
+.. autoclass:: transformers.modeling_tf_utils.TFTokenClassificationLoss
+    :members:
+
+
+``TensorFlow Helper Functions``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autofunction:: transformers.modeling_tf_utils.cast_bool_to_primitive
+
+.. autofunction:: transformers.modeling_tf_utils.get_initializer
+
+.. autofunction:: transformers.modeling_tf_utils.keras_serializable
+
+.. autofunction:: transformers.modeling_tf_utils.shape_list

docs/source/internal/pipelines_utils.rst

@@ -0,0 +1,40 @@
+Utilities for pipelines
+-----------------------
+
+This page lists all the utility functions the library provides for pipelines.
+
+Most of those are only useful if you are studying the code of the models in the library.
+
+
+Argument handling
+~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.pipelines.ArgumentHandler
+
+.. autoclass:: transformers.pipelines.ZeroShotClassificationArgumentHandler
+
+.. autoclass:: transformers.pipelines.QuestionAnsweringArgumentHandler
+
+
+Data format
+~~~~~~~~~~~
+
+.. autoclass:: transformers.pipelines.PipelineDataFormat
+    :members:
+
+.. autoclass:: transformers.pipelines.CsvPipelineDataFormat
+    :members:
+
+.. autoclass:: transformers.pipelines.JsonPipelineDataFormat
+    :members:
+
+.. autoclass:: transformers.pipelines.PipedPipelineDataFormat
+    :members:
+
+
+Utilities
+~~~~~~~~~
+
+.. autofunction:: transformers.pipelines.get_framework
+
+.. autoclass:: transformers.pipelines.PipelineException

docs/source/internal/tokenization_utils.rst

@@ -0,0 +1,38 @@
+Utilities for Tokenizers
+------------------------
+
+This page lists all the utility functions used by the tokenizers, mainly the class
+:class:`~transformers.tokenization_utils_base.PreTrainedTokenizerBase` that implements the common methods between
+:class:`~transformers.PreTrainedTokenizer` and :class:`~transformers.PreTrainedTokenizerFast` and the mixin
+:class:`~transformers.tokenization_utils_base.SpecialTokensMixin`.
+
+Most of those are only useful if you are studying the code of the tokenizers in the library.
+
+``PreTrainedTokenizerBase``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.tokenization_utils_base.PreTrainedTokenizerBase
+    :special-members: __call__
+    :members:
+
+
+``SpecialTokensMixin``
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.tokenization_utils_base.SpecialTokensMixin
+    :members:
+
+
+Enums and namedtuples
+~~~~~~~~~~~~~~~~~~~~~
+.. autoclass:: transformers.tokenization_utils_base.ExplicitEnum
+
+.. autoclass:: transformers.tokenization_utils_base.PaddingStrategy
+
+.. autoclass:: transformers.tokenization_utils_base.TensorType
+
+.. autoclass:: transformers.tokenization_utils_base.TruncationStrategy
+
+.. autoclass:: transformers.tokenization_utils_base.CharSpan
+
+.. autoclass:: transformers.tokenization_utils_base.TokenSpan

docs/source/main_classes/configuration.rst

@@ -1,7 +1,9 @@
 Configuration
 ----------------------------------------------------
 
-The base class ``PretrainedConfig`` implements the common methods for loading/saving a configuration either from a local file or directory, or from a pretrained model configuration provided by the library (downloaded from HuggingFace's AWS S3 repository).
+The base class ``PretrainedConfig`` implements the common methods for loading/saving a configuration either from a
+local file or directory, or from a pretrained model configuration provided by the library (downloaded from
+HuggingFace's AWS S3 repository).
 
 ``PretrainedConfig``
 ~~~~~~~~~~~~~~~~~~~~~

docs/source/main_classes/logging.rst

@@ -0,0 +1,44 @@
+Logging
+-------
+
+🤗 Transformers has a centralized logging system, so that you can setup the verbosity of the library easily. To
+change the level of verbosity, just use one of the direct setters. For instance, here is how to change the verbosity to
+the INFO level.
+
+.. code-block:: python
+
+    import transformers
+    transformers.logging.set_verbosity_info()
+
+All the methods of this logging module are documented below, the main ones are
+:func:`transformers.logging.get_verbosity` to get the current level of verbosity in the logger and
+:func:`transformers.logging.set_verbosity` to set the verbosity to the level of your choice. In order (from the least
+verbose to the most verbose), those levels (with their corresponding int values in parenthesis) are:
+
+- :obj:`transformers.logging.CRITICAL` or :obj:`transformers.logging.FATAL` (int value, 50): only report the most
+  critical errors.
+- :obj:`transformers.logging.ERROR` (int value, 40): only report errors.
+- :obj:`transformers.logging.WARNING` or :obj:`transformers.logging.WARN` (int value, 30): only reports error and
+  warnings. This the default level used by the library.
+- :obj:`transformers.logging.INFO` (int value, 20): reports error, warnings and basic information.
+- :obj:`transformers.logging.DEBUG` (int value, 10): report all information.
+
+Base setters
+~~~~~~~~~~~~
+
+.. autofunction:: transformers.logging.set_verbosity_error
+
+.. autofunction:: transformers.logging.set_verbosity_warning
+
+.. autofunction:: transformers.logging.set_verbosity_info
+
+.. autofunction:: transformers.logging.set_verbosity_debug
+
+Other functions
+~~~~~~~~~~~~~~~
+
+.. autofunction:: transformers.logging.get_verbosity
+
+.. autofunction:: transformers.logging.set_verbosity
+
+.. autofunction:: transformers.logging.get_logger

docs/source/main_classes/model.rst

@@ -1,23 +1,34 @@
 Models
 ----------------------------------------------------
 
-The base class ``PreTrainedModel`` implements the common methods for loading/saving a model either from a local file or directory, or from a pretrained model configuration provided by the library (downloaded from HuggingFace's AWS S3 repository).
+The base classes :class:`~transformers.PreTrainedModel` and :class:`~transformers.TFPreTrainedModel` implement the
+common methods for loading/saving a model either from a local file or directory, or from a pretrained model
+configuration provided by the library (downloaded from HuggingFace's AWS S3 repository).
 
-``PreTrainedModel`` also implements a few methods which are common among all the models to:
+:class:`~transformers.PreTrainedModel` and :class:`~transformers.TFPreTrainedModel` also implement a few methods which
+are common among all the models to:
 
 - resize the input token embeddings when new tokens are added to the vocabulary
 - prune the attention heads of the model.
 
+The other methods that are common to each model are defined in :class:`~transformers.modeling_utils.ModuleUtilsMixin`
+(for the PyTorch models) and :class:`~transformers.modeling_tf_utils.TFModuleUtilsMixin` (for the TensorFlow models) or
+for text generation, :class:`~transformers.generation_utils.GenerationMixin` (for the PyTorch models) and
+:class:`~transformers.generation_tf_utils.TFGenerationMixin` (for the TensorFlow models)
+
+
 ``PreTrainedModel``
 ~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.PreTrainedModel
     :members:
 
-``Helper Functions``
-~~~~~~~~~~~~~~~~~~~~~
 
-.. autofunction:: transformers.apply_chunking_to_forward
+``ModuleUtilsMixin``
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_utils.ModuleUtilsMixin
+    :members:
 
 
 ``TFPreTrainedModel``
@@ -25,3 +36,20 @@ The base class ``PreTrainedModel`` implements the common methods for loading/sav
 
 .. autoclass:: transformers.TFPreTrainedModel
     :members:
+
+
+``TFModelUtilsMixin``
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_tf_utils.TFModelUtilsMixin
+    :members:
+
+
+Generative models
+~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.generation_utils.GenerationMixin
+    :members:
+
+.. autoclass:: transformers.generation_tf_utils.TFGenerationMixin
+    :members:

docs/source/main_classes/optimizer_schedules.rst

@@ -1,4 +1,4 @@
-Optimizer
+Optimization
 ----------------------------------------------------
 
 The ``.optimization`` module provides:
@@ -7,24 +7,30 @@ The ``.optimization`` module provides:
 - several schedules in the form of schedule objects that inherit from ``_LRSchedule``:
 - a gradient accumulation class to accumulate the gradients of multiple batches
 
-``AdamW``
-~~~~~~~~~~~~~~~~
+``AdamW`` (PyTorch)
+~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.AdamW
     :members:
 
-``AdamWeightDecay``
-~~~~~~~~~~~~~~~~~~~
+``AdaFactor`` (PyTorch)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.Adafactor
+
+``AdamWeightDecay`` (TensorFlow)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.AdamWeightDecay
 
 .. autofunction:: transformers.create_optimizer
 
 Schedules
-----------------------------------------------------
+~~~~~~~~~~~~~~~~~~~
+
+Learning Rate Schedules (Pytorch)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-Learning Rate Schedules
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 .. autofunction:: transformers.get_constant_schedule
 
 
@@ -56,16 +62,16 @@ Learning Rate Schedules
     :target: /imgs/warmup_linear_schedule.png
     :alt:
 
-``Warmup``
-~~~~~~~~~~~~~~~~
+``Warmup`` (TensorFlow)
+^^^^^^^^^^^^^^^^^^^^^^^
 
 .. autoclass:: transformers.WarmUp
     :members:
 
 Gradient Strategies
-----------------------------------------------------
+~~~~~~~~~~~~~~~~~~~~
 
-``GradientAccumulator``
-~~~~~~~~~~~~~~~~~~~~~~~
+``GradientAccumulator`` (TensorFlow)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 .. autoclass:: transformers.GradientAccumulator

docs/source/main_classes/output.rst

@@ -0,0 +1,141 @@
+Model outputs
+-------------
+
+PyTorch models have outputs that are instances of subclasses of :class:`~transformers.file_utils.ModelOutput`. Those
+are data structures containing all the information returned by the model, but that can also be used as tuples or
+dictionaries.
+
+Let's see of this looks on an example:
+
+.. code-block::
+
+    from transformers import BertTokenizer, BertForSequenceClassification
+    import torch
+
+    tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+    model = BertForSequenceClassification.from_pretrained('bert-base-uncased')
+
+    inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
+    labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
+    outputs = model(**inputs, labels=labels)
+
+The ``outputs`` object is a :class:`~transformers.modeling_outputs.SequenceClassifierOutput`, as we can see in the
+documentation of that class below, it means it has an optional ``loss``, a ``logits`` an optional ``hidden_states`` and
+an optional ``attentions`` attribute. Here we have the ``loss`` since we passed along ``labels``, but we don't have
+``hidden_states`` and ``attentions`` because we didn't pass ``output_hidden_states=True`` or
+``output_attentions=True``.
+
+You can access each attribute as you would usually do, and if that attribute has not been returned by the model, you
+will get ``None``. Here for instance ``outputs.loss`` is the loss computed by the model, and ``outputs.attentions`` is
+``None``.
+
+When considering our ``outputs`` object as tuple, it only considers the attributes that don't have ``None`` values.
+Here for instance, it has two elements, ``loss`` then ``logits``, so
+
+.. code-block::
+
+    outputs[:2]
+
+will return the tuple ``(outputs.loss, outputs.logits)`` for instance.
+
+When considering our ``outputs`` object as dictionary, it only considers the attributes that don't have ``None``
+values. Here for instance, it has two keys that are ``loss`` and ``logits``.
+
+We document here the generic model outputs that are used by more than one model type. Specific output types are
+documented on their corresponding model page.
+
+``ModelOutput``
+~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.file_utils.ModelOutput
+    :members:
+
+``BaseModelOutput``
+~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.BaseModelOutput
+    :members:
+
+``BaseModelOutputWithPooling``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.BaseModelOutputWithPooling
+    :members:
+
+``BaseModelOutputWithPast``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.BaseModelOutputWithPast
+    :members:
+
+``Seq2SeqModelOutput``
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.Seq2SeqModelOutput
+    :members:
+
+``CausalLMOutput``
+~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.CausalLMOutput
+    :members:
+
+``CausalLMOutputWithPast``
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.CausalLMOutputWithPast
+    :members:
+
+``MaskedLMOutput``
+~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.MaskedLMOutput
+    :members:
+
+``Seq2SeqLMOutput``
+~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.Seq2SeqLMOutput
+    :members:
+
+``NextSentencePredictorOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.NextSentencePredictorOutput
+    :members:
+
+``SequenceClassifierOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.SequenceClassifierOutput
+    :members:
+
+``Seq2SeqSequenceClassifierOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.Seq2SeqSequenceClassifierOutput
+    :members:
+
+``MultipleChoiceModelOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.MultipleChoiceModelOutput
+    :members:
+
+``TokenClassifierOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.TokenClassifierOutput
+    :members:
+
+``QuestionAnsweringModelOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.QuestionAnsweringModelOutput
+    :members:
+
+``Seq2SeqQuestionAnsweringModelOutput``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_outputs.Seq2SeqQuestionAnsweringModelOutput
+    :members:

docs/source/main_classes/pipelines.rst

@@ -3,13 +3,24 @@ Pipelines
 
 The pipelines are a great and easy way to use models for inference. These pipelines are objects that abstract most
 of the complex code from the library, offering a simple API dedicated to several tasks, including Named Entity
-Recognition, Masked Language Modeling, Sentiment Analysis, Feature Extraction and Question Answering.
+Recognition, Masked Language Modeling, Sentiment Analysis, Feature Extraction and Question Answering. See the
+:doc:`task summary <../task_summary>` for examples of use.
 
 There are two categories of pipeline abstractions to be aware about:
 
-- The :func:`~transformers.pipeline` which is the most powerful object encapsulating all other pipelines
-- The other task-specific pipelines, such as :class:`~transformers.TokenClassificationPipeline`
-  or :class:`~transformers.QuestionAnsweringPipeline`
+- The :func:`~transformers.pipeline` which is the most powerful object encapsulating all other pipelines.
+- The other task-specific pipelines:
+
+    - :class:`~transformers.ConversationalPipeline`
+    - :class:`~transformers.FeatureExtractionPipeline`
+    - :class:`~transformers.FillMaskPipeline`
+    - :class:`~transformers.QuestionAnsweringPipeline`
+    - :class:`~transformers.SummarizationPipeline`
+    - :class:`~transformers.TextClassificationPipeline`
+    - :class:`~transformers.TextGenerationPipeline`
+    - :class:`~transformers.TokenClassificationPipeline`
+    - :class:`~transformers.TranslationPipeline`
+    - :class:`~transformers.ZeroShotClassificationPipeline`
 
 The pipeline abstraction
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -21,53 +32,82 @@ other pipeline but requires an additional argument which is the `task`.
 
 
 The task specific pipelines
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-Parent class: Pipeline
-=========================================
-
-.. autoclass:: transformers.Pipeline
-    :members: predict, transform, save_pretrained
-
-TokenClassificationPipeline
+ConversationalPipeline
 ==========================================
 
-.. autoclass:: transformers.TokenClassificationPipeline
+.. autoclass:: transformers.Conversation
 
-NerPipeline
-==========================================
-
-This class is an alias of the :class:`~transformers.TokenClassificationPipeline` defined above. Please refer to that pipeline for
-documentation and usage examples.
-
-FillMaskPipeline
-==========================================
-
-.. autoclass:: transformers.FillMaskPipeline
+.. autoclass:: transformers.ConversationalPipeline
+    :special-members: __call__
+    :members:
 
 FeatureExtractionPipeline
 ==========================================
 
 .. autoclass:: transformers.FeatureExtractionPipeline
+    :special-members: __call__
+    :members:
 
-TextClassificationPipeline
+FillMaskPipeline
 ==========================================
 
-.. autoclass:: transformers.TextClassificationPipeline
+.. autoclass:: transformers.FillMaskPipeline
+    :special-members: __call__
+    :members:
+
+NerPipeline
+==========================================
+
+This class is an alias of the :class:`~transformers.TokenClassificationPipeline` defined below. Please refer to that
+pipeline for documentation and usage examples.
 
 QuestionAnsweringPipeline
 ==========================================
 
 .. autoclass:: transformers.QuestionAnsweringPipeline
-
+    :special-members: __call__
+    :members:
 
 SummarizationPipeline
 ==========================================
 
 .. autoclass:: transformers.SummarizationPipeline
+    :special-members: __call__
+    :members:
 
+TextClassificationPipeline
+==========================================
+
+.. autoclass:: transformers.TextClassificationPipeline
+    :special-members: __call__
+    :members:
 
 TextGenerationPipeline
 ==========================================
 
 .. autoclass:: transformers.TextGenerationPipeline
+    :special-members: __call__
+    :members:
+
+TokenClassificationPipeline
+==========================================
+
+.. autoclass:: transformers.TokenClassificationPipeline
+    :special-members: __call__
+    :members:
+
+ZeroShotClassificationPipeline
+==========================================
+
+.. autoclass:: transformers.ZeroShotClassificationPipeline
+    :special-members: __call__
+    :members:
+
+
+Parent class: :obj:`Pipeline`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.Pipeline
+    :members:

docs/source/main_classes/tokenizer.rst

@@ -1,17 +1,40 @@
 Tokenizer
 ----------------------------------------------------
 
-A tokenizer is in charge of preparing the inputs for a model. The library comprise tokenizers for all the models. Most of the tokenizers are available in two flavors: a full python implementation and a "Fast" implementation based on the Rust library `tokenizers`. The "Fast" implementations allows (1) a significant speed-up in particular when doing batched tokenization and (2) additional methods to map between the original string (character and words) and the token space (e.g. getting the index of the token comprising a given character or the span of characters corresponding to a given token). Currently no "Fast" implementation is available for the SentencePiece-based tokenizers (for T5, ALBERT, CamemBERT, XLMRoBERTa and XLNet models).
+A tokenizer is in charge of preparing the inputs for a model. The library contains tokenizers for all the models. Most
+of the tokenizers are available in two flavors: a full python implementation and a "Fast" implementation based on the
+Rust library `tokenizers <https://github.com/huggingface/tokenizers>`__. The "Fast" implementations allows:
 
-The base classes ``PreTrainedTokenizer`` and ``PreTrainedTokenizerFast`` implements the common methods for encoding string inputs in model inputs (see below) and instantiating/saving python and "Fast" tokenizers either from a local file or directory or from a pretrained tokenizer provided by the library (downloaded from HuggingFace's AWS S3 repository).
+1. a significant speed-up in particular when doing batched tokenization and
+2. additional methods to map between the original string (character and words) and the token space (e.g. getting the
+   index of the token comprising a given character or the span of characters corresponding to a given token). Currently
+   no "Fast" implementation is available for the SentencePiece-based tokenizers (for T5, ALBERT, CamemBERT, XLMRoBERTa
+   and XLNet models).
 
-``PreTrainedTokenizer`` and ``PreTrainedTokenizerFast`` thus implements the main methods for using all the tokenizers:
+The base classes :class:`~transformers.PreTrainedTokenizer` and :class:`~transformers.PreTrainedTokenizerFast`
+implement the common methods for encoding string inputs in model inputs (see below) and instantiating/saving python and
+"Fast" tokenizers either from a local file or directory or from a pretrained tokenizer provided by the library
+(downloaded from HuggingFace's AWS S3 repository). They both rely on
+:class:`~transformers.tokenization_utils_base.PreTrainedTokenizerBase` that contains the common methods, and
+:class:`~transformers.tokenization_utils_base.SpecialTokensMixin`.
 
-- tokenizing (spliting strings in sub-word token strings), converting tokens strings to ids and back, and encoding/decoding (i.e. tokenizing + convert to integers),
-- adding new tokens to the vocabulary in a way that is independant of the underlying structure (BPE, SentencePiece...),
-- managing special tokens like mask, beginning-of-sentence, etc tokens (adding them, assigning them to attributes in the tokenizer for easy access and making sure they are not split during tokenization)
+:class:`~transformers.PreTrainedTokenizer` and :class:`~transformers.PreTrainedTokenizerFast` thus implement the main
+methods for using all the tokenizers:
+
+- Tokenizing (splitting strings in sub-word token strings), converting tokens strings to ids and back, and
+  encoding/decoding (i.e., tokenizing and converting to integers).
+- Adding new tokens to the vocabulary in a way that is independent of the underlying structure (BPE, SentencePiece...).
+- Managing special tokens (like mask, beginning-of-sentence, etc.): adding them, assigning them to attributes in the
+  tokenizer for easy access and making sure they are not split during tokenization.
+
+:class:`~transformers.BatchEncoding` holds the output of the tokenizer's encoding methods (``__call__``,
+``encode_plus`` and ``batch_encode_plus``) and is derived from a Python dictionary. When the tokenizer is a pure python
+tokenizer, this class behaves just like a standard python dictionary and holds the various model inputs computed by these
+methods (``input_ids``, ``attention_mask``...). When the tokenizer is a "Fast" tokenizer (i.e., backed by HuggingFace
+`tokenizers library <https://github.com/huggingface/tokenizers>`__), this class provides in addition several advanced
+alignment methods which can be used to map between the original string (character and words) and the token space (e.g.,
+getting the index of the token comprising a given character or the span of characters corresponding to a given token).
 
-``BatchEncoding`` holds the output of the tokenizer's encoding methods (``__call__``, ``encode_plus`` and ``batch_encode_plus``) and is derived from a Python dictionary. When the tokenizer is a pure python tokenizer, this class behave just like a standard python dictionary and hold the various model inputs computed by these methodes (``input_ids``, ``attention_mask``...). When the tokenizer is a "Fast" tokenizer (i.e. backed by HuggingFace tokenizers library), this class provides in addition several advanced alignement methods which can be used to map between the original string (character and words) and the token space (e.g. getting the index of the token comprising a given character or the span of characters corresponding to a given token).
 
 ``PreTrainedTokenizer``
 ~~~~~~~~~~~~~~~~~~~~~~~~
@@ -20,6 +43,7 @@ The base classes ``PreTrainedTokenizer`` and ``PreTrainedTokenizerFast`` impleme
     :special-members: __call__
     :members:
 
+
 ``PreTrainedTokenizerFast``
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -27,14 +51,9 @@ The base classes ``PreTrainedTokenizer`` and ``PreTrainedTokenizerFast`` impleme
     :special-members: __call__
     :members:
 
+
 ``BatchEncoding``
 ~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.BatchEncoding
     :members:
-
-``SpecialTokensMixin``
-~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.SpecialTokensMixin
-    :members:

docs/source/main_classes/trainer.rst

@@ -0,0 +1,62 @@
+Trainer
+----------
+
+The :class:`~transformers.Trainer` and :class:`~transformers.TFTrainer` classes provide an API for feature-complete
+training in most standard use cases. It's used in most of the :doc:`example scripts <../examples>`.
+
+Before instantiating your :class:`~transformers.Trainer`/:class:`~transformers.TFTrainer`, create a 
+:class:`~transformers.TrainingArguments`/:class:`~transformers.TFTrainingArguments` to access all the points of
+customization during training.
+
+The API supports distributed training on multiple GPUs/TPUs, mixed precision through `NVIDIA Apex
+<https://github.com/NVIDIA/apex>`__ for PyTorch and :obj:`tf.keras.mixed_precision` for TensorFlow.
+
+Both :class:`~transformers.Trainer` and :class:`~transformers.TFTrainer` contain the basic training loop supporting the
+previous features. To inject custom behavior you can subclass them and override the following methods:
+
+- **get_train_dataloader**/**get_train_tfdataset** -- Creates the training DataLoader (PyTorch) or TF Dataset.
+- **get_eval_dataloader**/**get_eval_tfdataset** -- Creates the evaulation DataLoader (PyTorch) or TF Dataset.
+- **get_test_dataloader**/**get_test_tfdataset** -- Creates the test DataLoader (PyTorch) or TF Dataset.
+- **log** -- Logs information on the various objects watching training.
+- **setup_wandb** -- Setups wandb (see `here <https://docs.wandb.com/huggingface>`__ for more information).
+- **create_optimizer_and_scheduler** -- Setups the optimizer and learning rate scheduler if they were not passed at
+  init.
+- **training_step** -- Performs a training step.
+- **prediction_step** -- Performs an evaluation/test step.
+- **run_model** (TensorFlow only) -- Basic pass through the model.
+- **evaluate** -- Runs an evaluation loop and returns metrics.
+- **predict** -- Returns predictions (with metrics if labels are available) on a test set.
+
+
+``Trainer`` 
+~~~~~~~~~~~
+
+.. autoclass:: transformers.Trainer
+    :members:
+
+``TFTrainer`` 
+~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFTrainer
+    :members:
+
+``TrainingArguments``
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TrainingArguments
+    :members:
+
+``TFTrainingArguments``
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFTrainingArguments
+    :members:
+
+Utilities
+~~~~~~~~~
+
+.. autoclass:: transformers.EvalPrediction
+
+.. autofunction:: transformers.set_seed
+
+.. autofunction:: transformers.torch_distributed_zero_first

docs/source/model_doc/albert.rst

@@ -47,6 +47,16 @@ AlbertTokenizer
         create_token_type_ids_from_sequences, save_vocabulary
 
 
+Albert specific outputs
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_albert.AlbertForPreTrainingOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_albert.TFAlbertForPreTrainingOutput
+    :members:
+
+
 AlbertModel
 ~~~~~~~~~~~~~~~~~~~~
 
@@ -54,6 +64,13 @@ AlbertModel
     :members:
 
 
+AlbertForPreTraining
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.AlbertForPreTraining
+    :members:
+
+
 AlbertForMaskedLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -96,6 +113,13 @@ TFAlbertModel
     :members:
 
 
+TFAlbertForPreTraining
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFAlbertForPreTraining
+    :members:
+
+
 TFAlbertForMaskedLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/bart.rst

@@ -17,13 +17,23 @@ According to the abstract,
 The Authors' code can be found `here <https://github.com/pytorch/fairseq/tree/master/examples/bart>`_
 
 
-Implementation Notes:
+Implementation Notes
+~~~~~~~~~~~~~~~~~~~~
 
 - Bart doesn't use :obj:`token_type_ids` for sequence classification. Use BartTokenizer.encode to get the proper splitting.
 - The forward pass of ``BartModel`` will create decoder inputs (using the helper function ``transformers.modeling_bart._prepare_bart_decoder_inputs``)  if they are not passed. This is different than some other modeling APIs.
 - Model predictions are intended to be identical to the original implementation. This only works, however, if the string you pass to ``fairseq.encode`` starts with a space.
 - ``BartForConditionalGeneration.generate`` should be used for conditional generation tasks like summarization, see the example in that docstrings
 - Models that load the ``"facebook/bart-large-cnn"`` weights will not have a ``mask_token_id``, or be able to perform mask filling tasks.
+- for training/forward passes that don't involve beam search, pass ``use_cache=False``
+
+
+BartForConditionalGeneration
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.BartForConditionalGeneration
+    :members: forward
+
 
 BartConfig
 ~~~~~~~~~~~~~~~~~~~~~
@@ -39,6 +49,7 @@ BartTokenizer
     :members:
 
 
+
 BartModel
 ~~~~~~~~~~~~~
 
@@ -62,10 +73,3 @@ BartForQuestionAnswering
     :members: forward
 
 
-BartForConditionalGeneration
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.BartForConditionalGeneration
-    :members: generate, forward
-
-

docs/source/model_doc/bert.rst

@@ -27,13 +27,8 @@ Tips:
 
 - BERT is a model with absolute position embeddings so it's usually advised to pad the inputs on
   the right rather than the left.
-- BERT was trained with a masked language modeling (MLM) objective. It is therefore efficient at predicting masked
-  tokens and at NLU in general, but is not optimal for text generation. Models trained with a causal language
-  modeling (CLM) objective are better in that regard.
-- Alongside MLM, BERT was trained using a next sentence prediction (NSP) objective using the [CLS] token as a sequence
-  approximate. The user may use this token (the first token in a sequence built with special tokens) to get a sequence
-  prediction rather than a token prediction. However, averaging over the sequence may yield better results than using
-  the [CLS] token.
+- BERT was trained with the masked language modeling (MLM) and next sentence prediction (NSP) objectives. It is efficient at predicting masked
+  tokens and at NLU in general, but is not optimal for text generation.
 
 The original code can be found `here <https://github.com/google-research/bert>`_.
 
@@ -59,6 +54,16 @@ BertTokenizerFast
     :members:
 
 
+Bert specific outputs
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_bert.BertForPreTrainingOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_bert.TFBertForPreTrainingOutput
+    :members:
+
+
 BertModel
 ~~~~~~~~~~~~~~~~~~~~
 
@@ -73,6 +78,13 @@ BertForPreTraining
     :members:
 
 
+BertModelLMHeadModel
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.BertLMHeadModel
+    :members:
+
+
 BertForMaskedLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -129,6 +141,13 @@ TFBertForPreTraining
     :members:
 
 
+TFBertModelLMHeadModel
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFBertLMHeadModel
+    :members:
+
+
 TFBertForMaskedLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/camembert.rst

@@ -49,6 +49,13 @@ CamembertModel
     :members:
 
 
+CamembertForCausalLM
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CamembertForCausalLM
+    :members:
+
+
 CamembertForMaskedLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -105,6 +112,13 @@ TFCamembertForSequenceClassification
     :members:
 
 
+TFCamembertForMultipleChoice
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFCamembertForMultipleChoice
+    :members:
+
+
 TFCamembertForTokenClassification
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/dpr.rst

@@ -0,0 +1,102 @@
+DPR
+----------------------------------------------------
+
+Overview
+~~~~~~~~~~~~~~~~~~~~~
+
+Dense Passage Retrieval (DPR) - is a set of tools and models for state-of-the-art open-domain Q&A research.
+It is based on the following paper:
+
+Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, Wen-tau Yih, Dense Passage Retrieval for Open-Domain Question Answering.
+
+The abstract from the paper is the following:
+
+*Open-domain question answering relies on efficient passage retrieval to select candidate contexts, where traditional
+sparse vector space models, such as TF-IDF or BM25, are the de facto method. In this work, we show that retrieval can
+be practically implemented using dense representations alone, where embeddings are learned from a small number of
+questions and passages by a simple dual-encoder framework. When evaluated on a wide range of open-domain QA datasets,
+our dense retriever outperforms a strong Lucene-BM25 system largely by 9%-19% absolute in terms of top-20 passage
+retrieval accuracy, and helps our end-to-end QA system establish new state-of-the-art on multiple open-domain QA
+benchmarks.*
+
+The original code can be found `here <https://github.com/facebookresearch/DPR>`_.
+
+
+DPRConfig
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRConfig
+    :members:
+
+
+DPRContextEncoderTokenizer
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRContextEncoderTokenizer
+    :members:
+
+
+DPRContextEncoderTokenizerFast
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRContextEncoderTokenizerFast
+    :members:
+
+DPRQuestionEncoderTokenizer
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRQuestionEncoderTokenizer
+    :members:
+
+
+DPRQuestionEncoderTokenizerFast
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRQuestionEncoderTokenizerFast
+    :members:
+
+DPRReaderTokenizer
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRReaderTokenizer
+    :members:
+
+
+DPRReaderTokenizerFast
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRReaderTokenizerFast
+    :members:
+
+
+DPR specific outputs
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_dpr.DPRContextEncoderOutput
+    :members:
+
+.. autoclass:: transformers.modeling_dpr.DPRQuestionEncoderOutput
+    :members:
+
+.. autoclass:: transformers.modeling_dpr.DPRReaderOutput
+    :members:
+
+
+DPRContextEncoder
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRContextEncoder
+    :members:
+
+DPRQuestionEncoder
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRQuestionEncoder
+    :members:
+
+
+DPRReader
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DPRReader
+    :members:

docs/source/model_doc/electra.rst

@@ -71,6 +71,16 @@ ElectraTokenizerFast
     :members:
 
 
+Electra specific outputs
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_electra.ElectraForPreTrainingOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_electra.TFElectraForPreTrainingOutput
+    :members:
+
+
 ElectraModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -99,6 +109,13 @@ ElectraForSequenceClassification
     :members:
 
 
+ElectraForMultipleChoice
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.ElectraForMultipleChoice
+    :members:
+
+
 ElectraForTokenClassification
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -134,6 +151,20 @@ TFElectraForMaskedLM
     :members:
 
 
+TFElectraForSequenceClassification
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFElectraForSequenceClassification
+    :members:
+
+
+TFElectraForMultipleChoice
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFElectraForMultipleChoice
+    :members:
+
+
 TFElectraForTokenClassification
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/encoderdecoder.rst

@@ -1,12 +1,13 @@
 Encoder Decoder Models
 ------------------------
 
-This class can wrap an encoder model, such as ``BertModel`` and a decoder modeling with a language modeling head, such as ``BertForMaskedLM`` into a encoder-decoder model.
+The :class:`~transformers.EncoderDecoderModel` can be used to initialize a sequence-to-sequence model with any pre-trained autoencoding model as the encoder and any pre-trained autoregressive model as the decoder.
 
-The ``EncoderDecoderModel`` class allows to instantiate a encoder decoder model using the ``from_encoder_decoder_pretrain`` class method taking a pretrained encoder and pretrained decoder model as an input. 
-The ``EncoderDecoderModel`` is saved using the standard ``save_pretrained()`` method and can also again be loaded using the standard ``from_pretrained()`` method. 
+The effectiveness of initializing sequence-to-sequence models with pre-trained checkpoints for sequence generation tasks was shown in `Leveraging Pre-trained Checkpoints for Sequence Generation Tasks <https://arxiv.org/abs/1907.12461>`__ by Sascha Rothe, Shashi Narayan, Aliaksei Severyn.
 
-An application of this architecture could be *summarization* using two pretrained Bert models as is shown in the paper: `Text Summarization with Pretrained Encoders <https://arxiv.org/abs/1910.13461>`_ by Yang Liu and Mirella Lapata. 
+After such an :class:`~transformers.EncoderDecoderModel` has been trained / fine-tuned, it can be saved / loaded just like any other models (see Examples for more information).
+
+An application of this architecture could be to leverage two pre-trained :obj:`transformers.BertModel` models as the encoder and decoder for a summarization model as was shown in: `Text Summarization with Pretrained Encoders <https://arxiv.org/abs/1910.13461>`_ by Yang Liu and Mirella Lapata. 
 
 
 ``EncoderDecoderConfig``

docs/source/model_doc/flaubert.rst

@@ -61,6 +61,20 @@ FlaubertForSequenceClassification
     :members:
 
 
+FlaubertForMultipleChoice
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.FlaubertForMultipleChoice
+    :members:
+
+
+FlaubertForTokenClassification
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.FlaubertForTokenClassification
+    :members:
+
+
 FlaubertForQuestionAnsweringSimple
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -114,4 +128,4 @@ TFFlaubertForQuestionAnsweringSimple
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.TFFlaubertForQuestionAnsweringSimple
-    :members:
+    :members:

docs/source/model_doc/gpt.rst

@@ -71,6 +71,16 @@ OpenAIGPTTokenizerFast
     :members:
 
 
+OpenAI specific outputs
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_openai.OpenAIGPTDoubleHeadsModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_openai.TFOpenAIGPTDoubleHeadsModelOutput
+    :members:
+
+
 OpenAIGPTModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/gpt2.rst

@@ -58,6 +58,16 @@ GPT2TokenizerFast
     :members:
 
 
+GPT2 specific outputs
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_gpt2.GPT2DoubleHeadsModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_gpt2.TFGPT2DoubleHeadsModelOutput
+    :members:
+
+
 GPT2Model
 ~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/longformer.rst

@@ -16,7 +16,7 @@ Longformer Self Attention
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 Longformer self attention employs self attention on both a "local" context and a "global" context.
 Most tokens only attend "locally" to each other meaning that each token attends to its :math:`\frac{1}{2} w` previous tokens and :math:`\frac{1}{2} w` succeding tokens with :math:`w` being the window length as defined in `config.attention_window`. Note that `config.attention_window` can be of type ``list`` to define a different :math:`w` for each layer. 
-A selecetd few tokens attend "globally" to all other tokens, as it is conventionally done for all tokens in *e.g.* `BertSelfAttention`.
+A selected few tokens attend "globally" to all other tokens, as it is conventionally done for all tokens in *e.g.* `BertSelfAttention`.
 
 Note that "locally" and "globally" attending tokens are projected by different query, key and value matrices.
 Also note that every "locally" attending token not only attends to tokens within its window :math:`w`, but also to all "globally" attending tokens so that global attention is *symmetric*.
@@ -102,3 +102,25 @@ LongformerForQuestionAnswering
 
 .. autoclass:: transformers.LongformerForQuestionAnswering
     :members:
+
+
+TFLongformerModel
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFLongformerModel
+    :members:
+
+
+TFLongformerForMaskedLM
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFLongformerForMaskedLM
+    :members:
+
+
+TFLongformerForQuestionAnswering
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.TFLongformerForQuestionAnswering
+    :members:
+

docs/source/model_doc/marian.rst

@@ -1,14 +1,14 @@
 MarianMT
 ----------------------------------------------------
-**DISCLAIMER:** If you see something strange,
-file a `Github Issue <https://github.com/huggingface/transformers/issues/new?assignees=&labels=&template=bug-report.md&title>`__ and assign
+**Bugs:** If you see something strange,
+file a `Github Issue <https://github.com/huggingface/transformers/issues/new?assignees=sshleifer&labels=&template=bug-report.md&title>`__ and assign
 @sshleifer. Translations should be similar, but not identical to, output in the test set linked to in each model card.
 
 Implementation Notes
 ~~~~~~~~~~~~~~~~~~~~
 - Each model is about 298 MB on disk, there are 1,000+ models.
 - The list of supported language pairs can be found `here <https://huggingface.co/Helsinki-NLP>`__.
-- The 1,000+ models were originally trained by `Jörg Tiedemann <https://researchportal.helsinki.fi/en/persons/j%C3%B6rg-tiedemann>`__ using the `Marian <https://marian-nmt.github.io/>`_ C++ library, which supports fast training and translation.
+- models were originally trained by `Jörg Tiedemann <https://researchportal.helsinki.fi/en/persons/j%C3%B6rg-tiedemann>`__ using the `Marian <https://marian-nmt.github.io/>`_ C++ library, which supports fast training and translation.
 - All models are transformer encoder-decoders with 6 layers in each component. Each model's performance is documented in a model card.
 - The 80 opus models that require BPE preprocessing are not supported.
 - The modeling code is the same as ``BartForConditionalGeneration`` with a few minor modifications:
@@ -48,7 +48,7 @@ Example of translating english to many romance languages, using language codes:
     tokenizer = MarianTokenizer.from_pretrained(model_name)
     print(tokenizer.supported_language_codes)
     model = MarianMTModel.from_pretrained(model_name)
-    translated = model.generate(**tokenizer.prepare_translation_batch(src_text))
+    translated = model.generate(**tokenizer.prepare_seq2seq_batch(src_text))
     tgt_text = [tokenizer.decode(t, skip_special_tokens=True) for t in translated]
     # ["c'est une phrase en anglais que nous voulons traduire en français",
     # 'Isto deve ir para o português.',
@@ -86,6 +86,14 @@ Code to see available pretrained models:
     suffix = [x.split('/')[1] for x in model_ids]
     multi_models = [f'{org}/{s}' for s in suffix if s != s.lower()]
 
+MarianMTModel
+~~~~~~~~~~~~~
+
+Pytorch version of marian-nmt's transformer.h (c++). Designed for the OPUS-NMT translation checkpoints.
+Model API is identical to BartForConditionalGeneration.
+Available models are listed at `Model List <https://huggingface.co/models?search=Helsinki-NLP>`__
+This class inherits nearly all functionality from ``BartForConditionalGeneration``, see that page for method signatures.
+
 MarianConfig
 ~~~~~~~~~~~~~~~~~~~
 .. autoclass:: transformers.MarianConfig
@@ -96,16 +104,8 @@ MarianTokenizer
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.MarianTokenizer
-    :members: prepare_translation_batch
+    :members: prepare_seq2seq_batch
 
 
-MarianMTModel
-~~~~~~~~~~~~~
 
-Pytorch version of marian-nmt's transformer.h (c++). Designed for the OPUS-NMT translation checkpoints.
-Model API is identical to BartForConditionalGeneration.
-Available models are listed at `Model List <https://huggingface.co/models?search=Helsinki-NLP>`__
-This class inherits all functionality from ``BartForConditionalGeneration``, see that page for method signatures.
 
-.. autoclass:: transformers.MarianMTModel
-    :members:

docs/source/model_doc/mbart.rst

@@ -0,0 +1,76 @@
+MBart
+----------------------------------------------------
+**DISCLAIMER:** If you see something strange,
+file a `Github Issue <https://github.com/huggingface/transformers/issues/new?assignees=&labels=&template=bug-report.md&title>`__ and assign
+@sshleifer
+
+Overview
+~~~~~~~~~~~~~~~~~~~~~
+The MBart model was presented in `Multilingual Denoising Pre-training for Neural Machine Translation <https://arxiv.org/abs/2001.08210>`_ by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov
+Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer. According to the abstract,
+
+MBART is a sequence-to-sequence denoising auto-encoder pre-trained on large-scale monolingual corpora in many languages using the BART objective. mBART is one of the first methods for pre-training a complete sequence-to-sequence model by denoising full texts in multiple languages, while previous approaches have focused only on the encoder, decoder, or reconstructing parts of the text.
+
+The Authors' code can be found `here <https://github.com/pytorch/fairseq/tree/master/examples/mbart>`__
+
+
+Training
+~~~~~~~~~~~~~~~~~~~~~
+MBart is a multilingual encoder-decoder (seq-to-seq) model primarily intended for translation task. 
+As the model is multilingual it expects the sequences in a different format. A special language id token 
+is added in both the source and target text. The source text format is ``X [eos, src_lang_code]`` 
+where ``X`` is the source text. The target text format is ```[tgt_lang_code] X [eos]```. ```bos``` is never used.
+The ```MBartTokenizer.prepare_seq2seq_batch``` handles this automatically and should be used to encode 
+the sequences for seq-2-seq fine-tuning.
+
+- Supervised training
+
+::
+
+    example_english_phrase = "UN Chief Says There Is No Military Solution in Syria"
+    expected_translation_romanian = "Şeful ONU declară că nu există o soluţie militară în Siria"
+    batch = tokenizer.prepare_seq2seq_batch(example_english_phrase, src_lang="en_XX", tgt_lang="ro_RO", tgt_texts=expected_translation_romanian)
+    input_ids = batch["input_ids"]
+    target_ids = batch["decoder_input_ids"]
+    decoder_input_ids = target_ids[:, :-1].contiguous()
+    labels = target_ids[:, 1:].clone()
+    model(input_ids=input_ids, decoder_input_ids=decoder_input_ids, labels=labels) #forward
+
+- Generation
+
+    While generating the target text set the `decoder_start_token_id` to the target language id. 
+    The following example shows how to translate English to Romanian using the ```facebook/mbart-large-en-ro``` model.
+
+::
+
+    from transformers import MBartForConditionalGeneration, MBartTokenizer
+    model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-en-ro")
+    tokenizer = MBartTokenizer.from_pretrained("facebook/mbart-large-en-ro")
+    article = "UN Chief Says There Is No Military Solution in Syria"
+    batch = tokenizer.prepare_seq2seq_batch(src_texts=[article], src_lang="en_XX")
+    translated_tokens = model.generate(**batch, decoder_start_token_id=tokenizer.lang_code_to_id["ro_RO"])
+    translation = tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
+    assert translation == "Şeful ONU declară că nu există o soluţie militară în Siria"
+
+
+MBartConfig
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.MBartConfig
+    :members:
+
+
+MBartTokenizer
+~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.MBartTokenizer
+    :members: build_inputs_with_special_tokens, prepare_seq2seq_batch
+
+
+MBartForConditionalGeneration
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.MBartForConditionalGeneration
+    :members: generate, forward
+
+

docs/source/model_doc/mobilebert.rst

@@ -56,6 +56,16 @@ MobileBertTokenizerFast
     :members:
 
 
+MobileBert specific outputs
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_mobilebert.MobileBertForPreTrainingOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_mobilebert.TFMobileBertForPreTrainingOutput
+    :members:
+
+
 MobileBertModel
 ~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/pegasus.rst

@@ -0,0 +1,117 @@
+Pegasus
+----------------------------------------------------
+**DISCLAIMER:** If you see something strange,
+file a `Github Issue <https://github.com/huggingface/transformers/issues/new?assignees=sshleifer&labels=&template=bug-report.md&title>`__ and assign
+@sshleifer.
+
+
+Overview
+~~~~~~~~~~~~~~~~~~~~~
+
+The Pegasus model was proposed in `PEGASUS: Pre-training with Extracted Gap-sentences for
+Abstractive Summarization <https://arxiv.org/pdf/1912.08777.pdf>`_ by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu on Dec 18, 2019.
+According to the abstract,
+
+- Pegasus' pretraining task is intentionally similar to summarization: important sentences are removed/masked from an input document and are generated together as one output sequence from the remaining sentences, similar to an extractive summary.
+- Pegasus achieves SOTA summarization performance on all 12 downstream tasks, as measured by ROUGE and human eval.
+
+The Authors' code can be found `here <https://github.com/google-research/pegasus>`_.
+
+
+Checkpoints
+~~~~~~~~~~~
+All the `checkpoints <https://huggingface.co/models?search=pegasus>`_ are finetuned for summarization, besides ``pegasus-large``, whence the other checkpoints are finetuned.
+- Each checkpoint is 2.2 GB on disk and 568M parameters.
+- FP16 is not supported (help/ideas on this appreciated!).
+- Summarizing xsum in fp32 takes about 400ms/sample, with default parameters on a v100 GPU.
+- For XSUM, The paper reports rouge1,rouge2, rougeL of paper: 47.21/24.56/39.25. As of Aug 9, this port scores 46.91/24.34/39.1.
+The gap is likely because of different alpha/length_penalty implementations in beam search.
+
+
+Implementation Notes
+~~~~~~~~~~~~~~~~~~~~
+
+- All models are transformer encoder-decoders with 16 layers in each component.
+- The implementation is completely inherited from ``BartForConditionalGeneration``
+- Some key configuration differences:
+    - static, sinusoidal position embeddings
+    - no ``layernorm_embedding`` (``PegasusConfig.normalize_embedding=False``)
+    - the model starts generating with pad_token_id (which has 0 token_embedding) as the prefix.
+    - ``num_beams=8``
+- All pretrained pegasus checkpoints are the same besides three attributes: ``tokenizer.model_max_length`` (max input size),  ``max_length`` (max num tokens to generate) and ``length_penalty``
+- Code to convert checkpoints trained in the author's `repo <https://github.com/google-research/pegasus>`_ can be found in ``convert_pegasus_tf_to_pytorch.py``
+
+
+Usage Example
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: python
+
+    from transformers import PegasusForConditionalGeneration, PegasusTokenizer
+    import torch
+    src_text = [
+        """ PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow."""
+    ]
+
+    model_name = 'google/pegasus-xsum'
+    torch_device = 'cuda' if torch.cuda.is_available() else 'cpu'
+    tokenizer = PegasusTokenizer.from_pretrained(model_name)
+    model = PegasusForConditionalGeneration.from_pretrained(model_name).to(torch_device)
+    batch = tokenizer.prepare_seq2seq_batch(src_text, truncation=True, padding='longest').to(torch_device)
+    translated = model.generate(**batch)
+    tgt_text = tokenizer.batch_decode(translated, skip_special_tokens=True)
+    assert tgt_text[0] == "California's largest electricity provider has turned off power to tens of thousands of customers."
+
+PegasusForConditionalGeneration
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+This class inherits all functionality from ``BartForConditionalGeneration``, see that page for method signatures.
+Available models are listed at `Model List <https://huggingface.co/models?search=pegasus>`__
+
+.. autoclass:: transformers.PegasusForConditionalGeneration
+    :members:
+
+
+PegasusConfig
+~~~~~~~~~~~~~~~~~~~
+This config fully inherits from ``BartConfig``, but pegasus uses different default values:
+Up to date parameter values can be seen in `S3 <https://s3.amazonaws.com/models.huggingface.co/bert/google/pegasus-xsum/config.json>`_.
+As of Aug 10, 2020, they are:
+
+.. code-block:: python
+
+    dict(
+    vocab_size=96103,
+    max_position_embeddings=512,
+    d_model=1024,
+    encoder_ffn_dim=4096,
+    decoder_ffn_dim=4096,
+    encoder_attention_heads=16,
+    decoder_attention_heads=16,
+    encoder_layers=16,
+    decoder_layers=16,
+    dropout=0.1,
+    attention_dropout=0.1,
+    activation_dropout=0.1,
+    pad_token_id=0,
+    eos_token_id=1,
+    is_encoder_decoder=True,
+    normalize_before=True,
+    scale_embedding=True,
+    normalize_embedding=False,
+    add_final_layer_norm=True,
+    static_position_embeddings=True,
+    num_beams=8,
+    activation_function="relu",
+    )
+
+
+PegasusTokenizer
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+warning: ``add_tokens`` does not work at the moment.
+
+.. autoclass:: transformers.PegasusTokenizer
+    :members: __call__, prepare_seq2seq_batch
+
+
+

docs/source/model_doc/reformer.rst

@@ -112,3 +112,24 @@ ReformerModelWithLMHead
 
 .. autoclass:: transformers.ReformerModelWithLMHead
     :members:
+
+
+ReformerForMaskedLM
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.ReformerForMaskedLM
+    :members:
+
+
+ReformerForSequenceClassification
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.ReformerForSequenceClassification
+    :members:
+
+
+ReformerForQuestionAnswering
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.ReformerForQuestionAnswering
+    :members:

docs/source/model_doc/roberta.rst

@@ -63,6 +63,13 @@ RobertaModel
     :members:
 
 
+RobertaForCausalLM
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.RobertaForCausalLM
+    :members:
+
+
 RobertaForMaskedLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/t5.rst

@@ -38,13 +38,13 @@ T5 can be trained / fine-tuned both in a supervised and unsupervised fashion.
 
   In this setup spans of the input sequence are masked by so-called sentinel tokens (*a.k.a* unique mask tokens) 
   and the output sequence is formed as a concatenation of the same sentinel tokens and the *real* masked tokens. 
-  Each sentinel token represents a unique mask token for this sentence and should start with ``<extra_id_1>``, ``<extra_id_2>``, ... up to ``<extra_id_100>``. As a default 100 sentinel tokens are available in ``T5Tokenizer``.
+  Each sentinel token represents a unique mask token for this sentence and should start with ``<extra_id_0>``, ``<extra_id_1>``, ... up to ``<extra_id_99>``. As a default 100 sentinel tokens are available in ``T5Tokenizer``.
   *E.g.* the sentence "The cute dog walks in the park" with the masks put on "cute dog" and "the" should be processed as follows: 
 
 ::
 
-  input_ids = tokenizer.encode('The <extra_id_1> walks in <extra_id_2> park', return_tensors='pt')
-  labels = tokenizer.encode('<extra_id_1> cute dog <extra_id_2> the <extra_id_3> </s>', return_tensors='pt')
+  input_ids = tokenizer.encode('The <extra_id_0> walks in <extra_id_1> park', return_tensors='pt')
+  labels = tokenizer.encode('<extra_id_0> cute dog <extra_id_1> the <extra_id_2> </s>', return_tensors='pt')
   # the forward function automatically creates the correct decoder_input_ids
   model(input_ids=input_ids, labels=labels)
 

docs/source/model_doc/transformerxl.rst

@@ -54,6 +54,22 @@ TransfoXLTokenizerFast
     :members:
 
 
+TransfoXL specific outputs
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_transfo_xl.TransfoXLModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_transfo_xl.TransfoXLLMHeadModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_transfo_xl.TFTransfoXLModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_transfo_xl.TFTransfoXLLMHeadModelOutput
+    :members:
+
+
 TransfoXLModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/xlm.rst

@@ -46,6 +46,14 @@ XLMTokenizer
     :members: build_inputs_with_special_tokens, get_special_tokens_mask,
         create_token_type_ids_from_sequences, save_vocabulary
 
+
+XLM specific outputs
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_xlm.XLMForQuestionAnsweringOutput
+    :members:
+
+
 XLMModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -67,6 +75,20 @@ XLMForSequenceClassification
     :members:
 
 
+XLMForMultipleChoice
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.XLMForMultipleChoice
+    :members:
+
+
+XLMForTokenClassification
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.XLMForTokenClassification
+    :members:
+
+
 XLMForQuestionAnsweringSimple
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_doc/xlnet.rst

@@ -50,6 +50,49 @@ XLNetTokenizer
         create_token_type_ids_from_sequences, save_vocabulary
 
 
+XLNet specific outputs
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.modeling_xlnet.XLNetModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_xlnet.XLNetLMHeadModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_xlnet.XLNetForSequenceClassificationOutput
+    :members:
+
+.. autoclass:: transformers.modeling_xlnet.XLNetForMultipleChoiceOutput
+    :members:
+
+.. autoclass:: transformers.modeling_xlnet.XLNetForTokenClassificationOutput
+    :members:
+
+.. autoclass:: transformers.modeling_xlnet.XLNetForQuestionAnsweringSimpleOutput
+    :members:
+
+.. autoclass:: transformers.modeling_xlnet.XLNetForQuestionAnsweringOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_xlnet.TFXLNetModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_xlnet.TFXLNetLMHeadModelOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_xlnet.TFXLNetForSequenceClassificationOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_xlnet.TFXLNetForMultipleChoiceOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_xlnet.TFXLNetForTokenClassificationOutput
+    :members:
+
+.. autoclass:: transformers.modeling_tf_xlnet.TFXLNetForQuestionAnsweringSimpleOutput
+    :members:
+
+
 XLNetModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/model_sharing.rst

@@ -171,8 +171,11 @@ Add a model card
 ^^^^^^^^^^^^^^^^
 
 To make sure everyone knows what your model can do, what its limitations and potential bias or ethetical
-considerations, please add a README.md model card to the 🤗 Transformers repo under `model_cards/`. It should be named
-`README.md` and follow `this template <https://github.com/huggingface/model_card>`__.
+considerations, please add a README.md model card to the 🤗 Transformers repo under `model_cards/`. It should then be
+placed in a subfolder with your username or organization, then another subfolder named like your model
+(`awesome-name-you-picked`). Or just click on the "Create a model card on GitHub" button on the model page, it will
+get you directly to the right location. If you need one, `here <https://github.com/huggingface/model_card>`__ is a
+model card template (meta-suggestions are welcome).
 
 If your model is fine-tuned from another model coming from the model hub (all 🤗 Transformers pretrained models do),
 don't forget to link to its model card so that people can fully trace how your model was built.
@@ -180,6 +183,11 @@ don't forget to link to its model card so that people can fully trace how your m
 If you have never made a pull request to the 🤗 Transformers repo, look at the
 :doc:`contributing guide <contributing>` to see the steps to follow.
 
+.. Note::
+
+    You can also send your model card in the folder you uploaded with the CLI by placing it in a `README.md` file
+    inside `path/to/awesome-name-you-picked/`.
+
 Using your model
 ^^^^^^^^^^^^^^^^
 

docs/source/model_summary.rst

@@ -1,11 +1,11 @@
 Summary of the models
 ================================================
 
-This is a summary of the models available in 🤗 Transformers. It assumes you’re familiar with the original 
-`transformer model <https://arxiv.org/abs/1706.03762>`_. For a gentle introduction check the `annotated transformer 
+This is a summary of the models available in 🤗 Transformers. It assumes you’re familiar with the original
+`transformer model <https://arxiv.org/abs/1706.03762>`_. For a gentle introduction check the `annotated transformer
 <http://nlp.seas.harvard.edu/2018/04/03/attention.html>`_. Here we focus on the high-level differences between the
-models. You can check them more in detail in their respective documentation. Also checkout the 
-:doc:`pretrained model page </pretrained_models>` to see the checkpoints available for each type of model and all `the 
+models. You can check them more in detail in their respective documentation. Also checkout the
+:doc:`pretrained model page </pretrained_models>` to see the checkpoints available for each type of model and all `the
 community models <https://huggingface.co/models>`_.
 
 Each one of the models in the library falls into one of the following categories:
@@ -14,38 +14,39 @@ Each one of the models in the library falls into one of the following categories
   * :ref:`autoencoding-models`
   * :ref:`seq-to-seq-models`
   * :ref:`multimodal-models`
+  * :ref:`retrieval-based-models`
 
-Autoregressive models are pretrained on the classic language modeling task: guess the next token having read all the 
-previous ones. They correspond to the decoder of the original transformer model, and a mask is used on top of the full 
-sentence so that the attention heads can only see what was before in the next, and not what’s after. Although those 
-models can be fine-tuned and achieve great results on many tasks, the most natural application is text generation. 
+Autoregressive models are pretrained on the classic language modeling task: guess the next token having read all the
+previous ones. They correspond to the decoder of the original transformer model, and a mask is used on top of the full
+sentence so that the attention heads can only see what was before in the next, and not what’s after. Although those
+models can be fine-tuned and achieve great results on many tasks, the most natural application is text generation.
 A typical example of such models is GPT.
 
-Autoencoding models are pretrained by corrupting the input tokens in some way and trying to reconstruct the original 
-sentence. They correspond to the encoder of the original transformer model in the sense that they get access to the 
-full inputs without any mask. Those models usually build a bidirectional representation of the whole sentence. They can 
-be fine-tuned and achieve great results on many tasks such as text generation, but their most natural application is 
+Autoencoding models are pretrained by corrupting the input tokens in some way and trying to reconstruct the original
+sentence. They correspond to the encoder of the original transformer model in the sense that they get access to the
+full inputs without any mask. Those models usually build a bidirectional representation of the whole sentence. They can
+be fine-tuned and achieve great results on many tasks such as text generation, but their most natural application is
 sentence classification or token classification. A typical example of such models is BERT.
 
-Note that the only difference between autoregressive models and autoencoding models is in the way the model is 
+Note that the only difference between autoregressive models and autoencoding models is in the way the model is
 pretrained. Therefore, the same architecture can be used for both autoregressive and autoencoding models. When a given
-model has been used for both pretraining, we have put it in the category corresponding to the article it was first
+model has been used for both types of pretraining, we have put it in the category corresponding to the article where it was first
 introduced.
 
-Sequence-to-sequence models use both the encoder and the decoder of the original transformer, either for translation 
-tasks or by transforming other tasks to sequence-to-sequence problems. They can be fine-tuned to many tasks but their 
-most natural applications are translation, summarization and question answering. The original transformer model is an 
+Sequence-to-sequence models use both the encoder and the decoder of the original transformer, either for translation
+tasks or by transforming other tasks to sequence-to-sequence problems. They can be fine-tuned to many tasks but their
+most natural applications are translation, summarization and question answering. The original transformer model is an
 example of such a model (only for translation), T5 is an example that can be fine-tuned on other tasks.
 
-Multimodal models mix text inputs with other kinds (like image) and are more specific to a given task.
+Multimodal models mix text inputs with other kinds (e.g. images) and are more specific to a given task.
 
 .. _autoregressive-models:
 
 Autoregressive models
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-As mentioned before, these models rely on the decoder part of the original transformer and use an attention mask so 
-that at each position, the model can only look at the tokens before in the attention heads.
+As mentioned before, these models rely on the decoder part of the original transformer and use an attention mask so
+that at each position, the model can only look at the tokens before the attention heads.
 
 Original GPT
 ----------------------------------------------
@@ -55,16 +56,16 @@ Original GPT
    <a href="https://huggingface.co/models?filter=openai-gpt">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-openai--gpt-blueviolet">
    </a>
-   <a href="/model_doc/gpt">
+   <a href="model_doc/gpt.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-openai--gpt-blueviolet">
    </a>
 
-`Improving Language Understanding by Generative Pre-Training <https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf>`_, 
+`Improving Language Understanding by Generative Pre-Training <https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf>`_,
 Alec Radford et al.
 
 The first autoregressive model based on the transformer architecture, pretrained on the Book Corpus dataset.
 
-The library provides versions of the model for language modeling and multitask language modeling/multiple choice 
+The library provides versions of the model for language modeling and multitask language modeling/multiple choice
 classification.
 
 GPT-2
@@ -75,17 +76,17 @@ GPT-2
    <a href="https://huggingface.co/models?filter=gpt2">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-gpt2-blueviolet">
    </a>
-   <a href="/model_doc/gpt2">
+   <a href="model_doc/gpt2.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-gpt2-blueviolet">
    </a>
 
-`Language Models are Unsupervised Multitask Learners <https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf>`_, 
+`Language Models are Unsupervised Multitask Learners <https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf>`_,
 Alec Radford et al.
 
-A bigger and better version of GPT, pretrained on WebText (web pages from outgoing links in Reddit with 3 karmas or 
+A bigger and better version of GPT, pretrained on WebText (web pages from outgoing links in Reddit with 3 karmas or
 more).
 
-The library provides versions of the model for language modeling and multitask language modeling/multiple choice 
+The library provides versions of the model for language modeling and multitask language modeling/multiple choice
 classification.
 
 CTRL
@@ -96,15 +97,15 @@ CTRL
    <a href="https://huggingface.co/models?filter=ctrl">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-ctrl-blueviolet">
    </a>
-   <a href="/model_doc/ctrl">
+   <a href="model_doc/ctrl.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-ctrl-blueviolet">
    </a>
 
-`CTRL: A Conditional Transformer Language Model for Controllable Generation <https://arxiv.org/abs/1909.05858>`_, 
+`CTRL: A Conditional Transformer Language Model for Controllable Generation <https://arxiv.org/abs/1909.05858>`_,
 Nitish Shirish Keskar et al.
 
-Same as the GPT model but adds the idea of control codes. Text is generated from a prompt (can be empty) and one (or 
-several) of those control codes which are then used to influence the text generation: generate with the style of 
+Same as the GPT model but adds the idea of control codes. Text is generated from a prompt (can be empty) and one (or
+several) of those control codes which are then used to influence the text generation: generate with the style of
 wikipedia article, a book or a movie review.
 
 The library provides a version of the model for language modeling only.
@@ -117,23 +118,23 @@ Transformer-XL
    <a href="https://huggingface.co/models?filter=transfo-xl">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-transfo--xl-blueviolet">
    </a>
-   <a href="/model_doc/transformerxl">
+   <a href="model_doc/transformerxl.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-transfo--xl-blueviolet">
    </a>
 
-`Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context <https://arxiv.org/abs/1901.02860>`_, 
+`Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context <https://arxiv.org/abs/1901.02860>`_,
 Zihang Dai et al.
 
-Same as a regular GPT model, but introduces a recurrence mechanism for two consecutive segments (similar to a regular 
-RNNs with two consecutive inputs). In this context, a segment is a number of consecutive tokens (for instance 512) that 
+Same as a regular GPT model, but introduces a recurrence mechanism for two consecutive segments (similar to a regular
+RNNs with two consecutive inputs). In this context, a segment is a number of consecutive tokens (for instance 512) that
 may span across multiple documents, and segments are fed in order to the model.
 
-Basically, the hidden states of the previous segment are concatenated to the current input to compute the attention 
-scores. This allows the model to pay attention to information that was in the previous segment as well as the current 
+Basically, the hidden states of the previous segment are concatenated to the current input to compute the attention
+scores. This allows the model to pay attention to information that was in the previous segment as well as the current
 one. By stacking multiple attention layers, the receptive field can be increased to multiple previous segments.
 
-This changes the positional embeddings to positional relative embeddings (as the regular positional embeddings would 
-give the same results in the current input and the current hidden state at a given position) and needs to make some 
+This changes the positional embeddings to positional relative embeddings (as the regular positional embeddings would
+give the same results in the current input and the current hidden state at a given position) and needs to make some
 adjustments in the way attention scores are computed.
 
 The library provides a version of the model for language modeling only.
@@ -148,23 +149,23 @@ Reformer
    <a href="https://huggingface.co/models?filter=reformer">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-reformer-blueviolet">
    </a>
-   <a href="/model_doc/reformer">
+   <a href="model_doc/reformer.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-reformer-blueviolet">
    </a>
 
 `Reformer: The Efficient Transformer <https://arxiv.org/abs/2001.04451>`_,
 Nikita Kitaev et al .
 
-An autoregressive transformer model with lots of tricks to reduce memory footprint and compute time. Those tricks 
+An autoregressive transformer model with lots of tricks to reduce memory footprint and compute time. Those tricks
 include:
 
-  * Use :ref:`Axial position encoding <axial-pos-encoding>` (see below for more details). It’s a mechanism to avoid 
-    having a huge positional encoding matrix (when the sequence length is very big) by factorizing it in smaller 
+  * Use :ref:`Axial position encoding <axial-pos-encoding>` (see below for more details). It’s a mechanism to avoid
+    having a huge positional encoding matrix (when the sequence length is very big) by factorizing it into smaller
     matrices.
-  * Replace traditional attention by :ref:`LSH (local-sensitive hashing) attention <lsh-attention>` (see below for more 
-    details). It's a technique to avoid compute the full product query-key in the attention layers.
-  * Avoid storing the intermediate results of each layer by using reversible transformer layers to obtain them during 
-    the backward pass (subtracting the residuals from the input of the next layer gives them back) or recomputing them 
+  * Replace traditional attention by :ref:`LSH (local-sensitive hashing) attention <lsh-attention>` (see below for more
+    details). It's a technique to avoid computing the full product query-key in the attention layers.
+  * Avoid storing the intermediate results of each layer by using reversible transformer layers to obtain them during
+    the backward pass (subtracting the residuals from the input of the next layer gives them back) or recomputing them
     for results inside a given layer (less efficient than storing them but saves memory).
   * Compute the feedforward operations by chunks and not on the whole batch.
 
@@ -183,21 +184,21 @@ XLNet
    <a href="https://huggingface.co/models?filter=xlnet">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-xlnet-blueviolet">
    </a>
-   <a href="/model_doc/xlnet">
+   <a href="model_doc/xlnet.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-xlnet-blueviolet">
    </a>
 
 `XLNet: Generalized Autoregressive Pretraining for Language Understanding <https://arxiv.org/abs/1906.08237>`_,
 Zhilin Yang et al.
 
-XLNet is not a traditional autoregressive model but uses a training strategy that builds on that. It permutes the 
-tokens in the sentence, then allows the model to use the last n tokens to predict the token n+1. Since this is all done 
-with a mask, the sentence is actually fed in the model in the right order, but instead of masking the first n tokens 
+XLNet is not a traditional autoregressive model but uses a training strategy that builds on that. It permutes the
+tokens in the sentence, then allows the model to use the last n tokens to predict the token n+1. Since this is all done
+with a mask, the sentence is actually fed in the model in the right order, but instead of masking the first n tokens
 for n+1, XLNet uses a mask that hides the previous tokens in some given permutation of 1,...,sequence length.
 
-XLNet also uses the same recurrence mechanism as TransformerXL to build long-term dependencies. 
+XLNet also uses the same recurrence mechanism as Transformer-XL to build long-term dependencies.
 
-The library provides a version of the model for language modeling, token classification, sentence classification, 
+The library provides a version of the model for language modeling, token classification, sentence classification,
 multiple choice classification and question answering.
 
 .. _autoencoding-models:
@@ -206,8 +207,7 @@ Autoencoding models
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 As mentioned before, these models rely on the encoder part of the original transformer and use no mask so the model can
-look at all the tokens in the attention heads. For pretraining, inputs are a corrupted version of the sentence, usually 
-obtained by masking tokens, and targets are the original sentences.
+look at all the tokens in the attention heads. For pretraining, targets are the original sentences and inputs are their corrupted versions.
 
 BERT
 ----------------------------------------------
@@ -217,25 +217,25 @@ BERT
    <a href="https://huggingface.co/models?filter=bert">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-bert-blueviolet">
    </a>
-   <a href="/model_doc/bert">
+   <a href="model_doc/bert.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-bert-blueviolet">
    </a>
 
 `BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding <https://arxiv.org/abs/1810.04805>`_,
 Jacob Devlin et al.
 
-Corrupts the inputs by using random masking, more precisely, during pretraining, a given percentage of tokens (usually 
-15%) are masked by
- 
+Corrupts the inputs by using random masking, more precisely, during pretraining, a given percentage of tokens (usually
+15%) is masked by:
+
   * a special mask token with probability 0.8
   * a random token different from the one masked with probability 0.1
   * the same token with probability 0.1
 
-The model must predict the original sentence, but has a second objective: inputs are two sentences A and B (with a 
-separation token in between). With probability 50%, the sentences are consecutive in the corpus, in the remaining 50% 
+The model must predict the original sentence, but has a second objective: inputs are two sentences A and B (with a
+separation token in between). With probability 50%, the sentences are consecutive in the corpus, in the remaining 50%
 they are not related. The model has to predict if the sentences are consecutive or not.
 
-The library provides a version of the model for language modeling (traditional or masked), next sentence prediction, 
+The library provides a version of the model for language modeling (traditional or masked), next sentence prediction,
 token classification, sentence classification, multiple choice classification and question answering.
 
 ALBERT
@@ -246,7 +246,7 @@ ALBERT
    <a href="https://huggingface.co/models?filter=albert">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-albert-blueviolet">
    </a>
-   <a href="/model_doc/albert">
+   <a href="model_doc/albert.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-albert-blueviolet">
    </a>
 
@@ -255,16 +255,16 @@ Zhenzhong Lan et al.
 
 Same as BERT but with a few tweaks:
 
-  * Embedding size E is different from hidden size H justified because the embeddings are context independent (one 
-    embedding vector represents one token) whereas hidden states are context dependent (one hidden state represents a 
-    sequence of tokens) so it's more logical to have H >> E. Als, the embedding matrix is large since it's V x E (V 
+  * Embedding size E is different from hidden size H justified because the embeddings are context independent (one
+    embedding vector represents one token), whereas hidden states are context dependent (one hidden state represents a
+    sequence of tokens) so it's more logical to have H >> E. Also, the embedding matrix is large since it's V x E (V
     being the vocab size). If E < H, it has less parameters.
   * Layers are split in groups that share parameters (to save memory).
-  * Next sentence prediction is replaced by a sentence ordering prediction: in the inputs, we have two sentences A et B 
-    (that are consecutive) and we either feed A followed by B or B followed by A. The model must predict if they have 
+  * Next sentence prediction is replaced by a sentence ordering prediction: in the inputs, we have two sentences A and B
+    (that are consecutive) and we either feed A followed by B or B followed by A. The model must predict if they have
     been swapped or not.
 
-The library provides a version of the model for masked language modeling, token classification, sentence 
+The library provides a version of the model for masked language modeling, token classification, sentence
 classification, multiple choice classification and question answering.
 
 RoBERTa
@@ -275,7 +275,7 @@ RoBERTa
    <a href="https://huggingface.co/models?filter=roberta">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-roberta-blueviolet">
    </a>
-   <a href="/model_doc/roberta">
+   <a href="model_doc/roberta.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-roberta-blueviolet">
    </a>
 
@@ -284,13 +284,13 @@ Yinhan Liu et al.
 
 Same as BERT with better pretraining tricks:
 
-  * dynamic masking: tokens are masked differently at each epoch whereas BERT does it once and for all
-  * no NSP (next sentence prediction) loss and instead of putting just two sentences together, put a chunk of 
-    contiguous texts together to reach 512 tokens (so sentences in in an order than may span other several documents)
+  * dynamic masking: tokens are masked differently at each epoch, whereas BERT does it once and for all
+  * no NSP (next sentence prediction) loss and instead of putting just two sentences together, put a chunk of
+    contiguous texts together to reach 512 tokens (so the sentences are in an order than may span several documents)
   * train with larger batches
   * use BPE with bytes as a subunit and not characters (because of unicode characters)
 
-The library provides a version of the model for masked language modeling, token classification, sentence 
+The library provides a version of the model for masked language modeling, token classification, sentence
 classification, multiple choice classification and question answering.
 
 DistilBERT
@@ -301,21 +301,21 @@ DistilBERT
    <a href="https://huggingface.co/models?filter=distilbert">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-distilbert-blueviolet">
    </a>
-   <a href="/model_doc/distilbert">
+   <a href="model_doc/distilbert.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-distilbert-blueviolet">
    </a>
 
 `DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter <https://arxiv.org/abs/1910.01108>`_,
 Victor Sanh et al.
 
-Same as BERT but smaller. Trained by distillation of the pretrained BERT model, meaning it's been trained to predict 
+Same as BERT but smaller. Trained by distillation of the pretrained BERT model, meaning it's been trained to predict
 the same probabilities as the larger model. The actual objective is a combination of:
 
   * finding the same probabilities as the teacher model
   * predicting the masked tokens correctly (but no next-sentence objective)
   * a cosine similarity between the hidden states of the student and the teacher model
 
-The library provides a version of the model for masked language modeling, token classification, sentence classification 
+The library provides a version of the model for masked language modeling, token classification, sentence classification
 and question answering.
 
 XLM
@@ -326,31 +326,30 @@ XLM
    <a href="https://huggingface.co/models?filter=xlm">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-xlm-blueviolet">
    </a>
-   <a href="/model_doc/xlm">
+   <a href="model_doc/xlm.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-xlm-blueviolet">
    </a>
 
 `Cross-lingual Language Model Pretraining <https://arxiv.org/abs/1901.07291>`_, Guillaume Lample and Alexis Conneau
 
-A transformer model trained on several languages. There are three different type of training for this model and the 
+A transformer model trained on several languages. There are three different type of training for this model and the
 library provides checkpoints for all of them:
 
-  * Causal language modeling (CLM) which is the traditional autoregressive training (so this model could be in the 
-    previous section as well). One of the languages is selected for each training sample, and the model input is a 
-    sentence of 256 tokens that may span on several documents in one one those languages.
-  * Masked language modeling (MLM) which is like RoBERTa. One of the languages is selected for each training sample, 
-    and the model input is a sentence of 256 tokens that may span on several documents in one one those languages, with
+  * Causal language modeling (CLM) which is the traditional autoregressive training (so this model could be in the
+    previous section as well). One of the languages is selected for each training sample, and the model input is a
+    sentence of 256 tokens, that may span over several documents in one of those languages.
+  * Masked language modeling (MLM) which is like RoBERTa. One of the languages is selected for each training sample,
+    and the model input is a sentence of 256 tokens, that may span over several documents in one of those languages, with
     dynamic masking of the tokens.
-  * A combination of MLM and translation language modeling (TLM). This consists of concatenating a sentence in two 
-    different languages, with random masking. To predict one of the masked token, the model can use both the 
-    surrounding context in language 1 as well as the context given by language 2.
+  * A combination of MLM and translation language modeling (TLM). This consists of concatenating a sentence in two
+    different languages, with random masking. To predict one of the masked tokens, the model can use both, the
+    surrounding context in language 1 and the context given by language 2.
 
 Checkpoints refer to which method was used for pretraining by having `clm`, `mlm` or `mlm-tlm` in their names. On top
 of positional embeddings, the model has language embeddings. When training using MLM/CLM, this gives the model an
-indication of the language used, and when training using MLM+TLM, an indication of which part of the input is in which
-language.
+indication of the language used, and when training using MLM+TLM, an indication of the language used for each part.
 
-The library provides a version of the model for language modeling, token classification, sentence classification and 
+The library provides a version of the model for language modeling, token classification, sentence classification and
 question answering.
 
 XLM-RoBERTa
@@ -361,18 +360,18 @@ XLM-RoBERTa
    <a href="https://huggingface.co/models?filter=xlm-roberta">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-xlm--roberta-blueviolet">
    </a>
-   <a href="/model_doc/xlmroberta">
+   <a href="model_doc/xlmroberta.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-xlm--roberta-blueviolet">
    </a>
 
-`Unsupervised Cross-lingual Representation Learning at Scale <https://arxiv.org/abs/1911.02116>`_, Alexis Conneau et 
+`Unsupervised Cross-lingual Representation Learning at Scale <https://arxiv.org/abs/1911.02116>`_, Alexis Conneau et
 al.
 
-Uses RoBERTa tricks on the XLM approach, but does not use the translation language modeling objective, only using 
-masked language modeling on sentences coming from one language. However, the model is trained on many more languages 
+Uses RoBERTa tricks on the XLM approach, but does not use the translation language modeling objective. It only uses
+masked language modeling on sentences coming from one language. However, the model is trained on many more languages
 (100) and doesn't use the language embeddings, so it's capable of detecting the input language by itself.
 
-The library provides a version of the model for masked language modeling, token classification, sentence 
+The library provides a version of the model for masked language modeling, token classification, sentence
 classification, multiple choice classification and question answering.
 
 FlauBERT
@@ -383,7 +382,7 @@ FlauBERT
    <a href="https://huggingface.co/models?filter=flaubert">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-flaubert-blueviolet">
    </a>
-   <a href="/model_doc/flaubert">
+   <a href="model_doc/flaubert.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-flaubert-blueviolet">
    </a>
 
@@ -401,20 +400,20 @@ ELECTRA
    <a href="https://huggingface.co/models?filter=electra">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-electra-blueviolet">
    </a>
-   <a href="/model_doc/electra">
+   <a href="model_doc/electra.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-electra-blueviolet">
    </a>
 
-`ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators <https://arxiv.org/abs/2003.10555>`_, 
+`ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators <https://arxiv.org/abs/2003.10555>`_,
 Kevin Clark et al.
 
-ELECTRA is a transformer model pretrained with the use of another (small) masked language model. The inputs are 
-corrupted by that language model, which takes an input text that is randomly masked and outputs a text in which ELECTRA 
-has to predict which token is an original and which one has been replaced. Like for GAN training, the small language 
-model is trained for a few steps (but with the original texts as objective, not to fool the ELECTRA model like in a 
+ELECTRA is a transformer model pretrained with the use of another (small) masked language model. The inputs are
+corrupted by that language model, which takes an input text that is randomly masked and outputs a text in which ELECTRA
+has to predict which token is an original and which one has been replaced. Like for GAN training, the small language
+model is trained for a few steps (but with the original texts as objective, not to fool the ELECTRA model like in a
 traditional GAN setting) then the ELECTRA model is trained for a few steps.
 
-The library provides a version of the model for masked language modeling, token classification and sentence 
+The library provides a version of the model for masked language modeling, token classification and sentence
 classification.
 
 .. _longformer:
@@ -427,15 +426,15 @@ Longformer
    <a href="https://huggingface.co/models?filter=longformer">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-longformer-blueviolet">
    </a>
-   <a href="/model_doc/longformer">
+   <a href="model_doc/longformer.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-longformer-blueviolet">
    </a>
 
 `Longformer: The Long-Document Transformer <https://arxiv.org/abs/2004.05150>`_, Iz Beltagy et al.
 
-A transformer model replacing the attention matrices by sparse matrices to go faster. Often, the local context (e.g., 
-what are the two tokens left and right?) is enough to take action for a given token. Some preselected input tokens are 
-still given global attention, but the attention matrix has way less parameters, resulting in a speed-up. See the 
+A transformer model replacing the attention matrices by sparse matrices to go faster. Often, the local context (e.g.,
+what are the two tokens left and right?) is enough to take action for a given token. Some preselected input tokens are
+still given global attention, but the attention matrix has way less parameters, resulting in a speed-up. See the
 :ref:`local attention section <local-attention>` for more information.
 
 It is pretrained the same way a RoBERTa otherwise.
@@ -443,7 +442,7 @@ It is pretrained the same way a RoBERTa otherwise.
 **Note:** This model could be very well be used in an autoregressive setting, there is no checkpoint for such a
 pretraining yet, though.
 
-The library provides a version of the model for masked language modeling, token classification, sentence 
+The library provides a version of the model for masked language modeling, token classification, sentence
 classification, multiple choice classification and question answering.
 
 .. _seq-to-seq-models:
@@ -461,25 +460,50 @@ BART
    <a href="https://huggingface.co/models?filter=bart">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-bart-blueviolet">
    </a>
-   <a href="/model_doc/bart">
+   <a href="model_doc/bart.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-bart-blueviolet">
    </a>
 
-`BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension 
+`BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
 <https://arxiv.org/abs/1910.13461>`_, Mike Lewis et al.
 
-Sequence-to-sequence model with an encoder and a decoder. Encoder is fed a corrupted version of the tokens, decoder is 
-fed the tokens (but has a mask to hide the future words like a regular transformers decoder). For the encoder, on the 
+Sequence-to-sequence model with an encoder and a decoder. Encoder is fed a corrupted version of the tokens, decoder is
+fed the original tokens (but has a mask to hide the future words like a regular transformers decoder). For the encoder, on the
 pretraining tasks, a composition of the following transformations are applied:
 
   * mask random tokens (like in BERT)
   * delete random tokens
   * mask a span of k tokens with a single mask token (a span of 0 tokens is an insertion of a mask token)
   * permute sentences
-  * rotate the document to make it start by a specific token
+  * rotate the document to make it start at a specific token
 
 The library provides a version of this model for conditional generation and sequence classification.
 
+Pegasus
+----------------------------------------------
+
+.. raw:: html
+
+   <a href="https://huggingface.co/models?filter=pegasus">
+       <img alt="Models" src="https://img.shields.io/badge/All_model_pages-pegasus-blueviolet">
+   </a>
+   <a href="model_doc/pegasus.html">
+       <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-pegasus-blueviolet">
+   </a>
+
+`PEGASUS: Pre-training with Extracted Gap-sentences forAbstractive Summarization 
+<https://arxiv.org/pdf/1912.08777.pdf>`_, Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu on Dec 18, 2019.
+
+Sequence-to-sequence model with the same encoder-decoder model architecture as BART. Pegasus is pre-trained jointly on two self-supervised objective functions: Masked Language Modeling (MLM) and a novel summarization specific pre-training objective, called Gap Sentence Generation (GSG).
+
+  * MLM: encoder input tokens are randomely replaced by a mask tokens and have to be predicted by the encoder (like in BERT)
+  * GSG: whole encoder input sentences are replaced by a second mask token and fed to the decoder, but which has a causal mask to hide the future words like a regular auto-regressive transformer decoder.
+
+In contrast to BART, Pegasus' pretraining task is intentionally similar to summarization: important sentences are masked and are generated together as one output sequence from the remaining sentences, similar to an extractive summary.
+
+The library provides a version of this model for conditional generation, which should be used for summarization.
+
+
 MarianMT
 ----------------------------------------------
 
@@ -488,7 +512,7 @@ MarianMT
    <a href="https://huggingface.co/models?filter=marian">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-marian-blueviolet">
    </a>
-   <a href="/model_doc/marian">
+   <a href="model_doc/marian.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-marian-blueviolet">
    </a>
 
@@ -506,97 +530,155 @@ T5
    <a href="https://huggingface.co/models?filter=t5">
        <img alt="Models" src="https://img.shields.io/badge/All_model_pages-t5-blueviolet">
    </a>
-   <a href="/model_doc/t5">
+   <a href="model_doc/t5.html">
        <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-t5-blueviolet">
    </a>
 
-`Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer <https://arxiv.org/abs/1910.10683>`_, 
+`Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer <https://arxiv.org/abs/1910.10683>`_,
 Colin Raffel et al.
 
-Uses the traditional transformer model (except a slight change with the positional embeddings, which are learned at 
-each layer). To be able to operate on all NLP tasks, it transforms them in text-to-text problems by using certain 
-prefixes: “Summarize: …”, “question: …”, “translate English to German: …” and so forth.
+Uses the traditional transformer model (with a slight change in the positional embeddings, which are learned at
+each layer). To be able to operate on all NLP tasks, it transforms them into text-to-text problems by using specific
+prefixes: “summarize: ”, “question: ”, “translate English to German: ” and so forth.
 
-The pretraining includes both supervised and self-supervised training. Supervised training is conducted on downstream 
-tasks provided by the GLUE and SuperGLUE benchmarks (changing them to text-to-text tasks as explained above).
+The pretraining includes both supervised and self-supervised training. Supervised training is conducted on downstream
+tasks provided by the GLUE and SuperGLUE benchmarks (converting them into text-to-text tasks as explained above).
 
-Self-supervised training consists of corrupted pretrained, which means randomly removing 15% of the tokens and 
-replacing them by individual sentinel tokens (if several consecutive tokens are marked for removal, they are replaced 
-by one single sentinel token). The input of the encoder is the corrupted sentence, the input of the decoder the 
+Self-supervised training uses corrupted tokens, by randomly removing 15% of the tokens and
+replacing them with individual sentinel tokens (if several consecutive tokens are marked for removal, the whole group is replaced with a single sentinel token). The input of the encoder is the corrupted sentence, the input of the decoder is the
 original sentence and the target is then the dropped out tokens delimited by their sentinel tokens.
 
-For instance, if we have the sentence “My dog is very cute .”, and we decide to remove the token dog, is and cute, the 
-input becomes “My <x> very <y> .” and the target is “<x> dog is <y> . <z>”
+For instance, if we have the sentence “My dog is very cute .”, and we decide to remove the tokens: "dog", "is" and "cute", the encoder
+input becomes “My <x> very <y> .” and the target input becomes “<x> dog is <y> cute .<z>”
 
 The library provides a version of this model for conditional generation.
 
+MBart
+----------------------------------------------
+
+.. raw:: html
+
+   <a href="https://huggingface.co/models?filter=mbart">
+       <img alt="Models" src="https://img.shields.io/badge/All_model_pages-mbart-blueviolet">
+   </a>
+   <a href="model_doc/mbart.html">
+       <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-mbart-blueviolet">
+   </a>
+
+`Multilingual Denoising Pre-training for Neural Machine Translation <https://arxiv.org/abs/2001.08210>`_ by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov
+Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer.
+
+The model architecture and pre-training objective is same as BART, but MBart is trained on 25 languages 
+and is intended for supervised and unsupervised machine translation. MBart is one of the first methods 
+for pre-training a complete sequence-to-sequence model by denoising full texts in multiple languages,
+
+The library provides a version of this model for conditional generation.
+
+The `mbart-large-en-ro checkpoint <https://huggingface.co/facebook/mbart-large-en-ro>`_ can be used for english -> romanian translation.
+
+The `mbart-large-cc25 <https://huggingface.co/facebook/mbart-large-cc25>`_ checkpoint can be finetuned for other translation and summarization tasks, using code in ```examples/seq2seq/``` , but is not very useful without finetuning.
+
 .. _multimodal-models:
 
 Multimodal models
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-There is one multimodal model in the library which has not been pretrained in the self-supervised fashion like the 
+There is one multimodal model in the library which has not been pretrained in the self-supervised fashion like the
 others.
 
 MMBT
 ----------------------------------------------
 
-`Supervised Multimodal Bitransformers for Classifying Images and Text <https://arxiv.org/abs/1909.02950>`_, Douwe Kiela 
+`Supervised Multimodal Bitransformers for Classifying Images and Text <https://arxiv.org/abs/1909.02950>`_, Douwe Kiela
 et al.
 
-A transformers model used in multimodal settings, combining a text and an image to make predictions. The transformer 
-model takes as inputs the embeddings of the tokenized text and a the final activations of a pretrained resnet on the 
-images (after the pooling layer) that goes through a linear layer (to go from number of features at the end of the 
+A transformers model used in multimodal settings, combining a text and an image to make predictions. The transformer
+model takes as inputs the embeddings of the tokenized text and the final activations of a pretrained on images resnet
+(after the pooling layer) that goes through a linear layer (to go from number of features at the end of the
 resnet to the hidden state dimension of the transformer).
 
-The different inputs are concatenated, and on top of the positional embeddings, a segment embedding is added to let the 
-model know which part of the input vector corresponds to the text or the image.
+The different inputs are concatenated, and on top of the positional embeddings, a segment embedding is added to let the
+model know which part of the input vector corresponds to the text and which to the image.
 
 The pretrained model only works for classification.
 
 ..
-    More information in this :doc:`model documentation </model_doc/mmbt>`.
+    More information in this :doc:`model documentation </model_doc/mmbt.html>`.
     TODO: write this page
 
+.. _retrieval-based-models:
+
+Retrieval-based models
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Some models use documents retrieval during (pre)training and inference for open-domain question answering, for example.
+
+
+DPR
+----------------------------------------------
+
+.. raw:: html
+
+   <a href="https://huggingface.co/models?filter=dpr">
+       <img alt="Models" src="https://img.shields.io/badge/All_model_pages-dpr-blueviolet">
+   </a>
+   <a href="model_doc/ctrl.dpr">
+       <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-dpr-blueviolet">
+   </a>
+
+`Dense Passage Retrieval for Open-Domain Question Answering <https://arxiv.org/abs/2004.04906>`_,
+Vladimir Karpukhin et al.
+
+Dense Passage Retrieval (DPR) - is a set of tools and models for state-of-the-art open-domain question-answering research.
+
+
+DPR consists in three models:
+
+  * Question encoder: encode questions as vectors
+  * Context encoder: encode contexts as vectors
+  * Reader: extract the answer of the questions inside retrieved contexts, along with a relevance score (high if the inferred span actually answers the question).
+
+DPR's pipeline (not implemented yet) uses a retrieval step to find the top k contexts given a certain question, and then it calls the reader with the question and the retrieved documents to get the answer.
+
 More technical aspects
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Full vs sparse attention
 ----------------------------------------------
 
-Most transformer models use full attention in the sense that the attention matrix is square. It can be a big 
-computational bottleneck when you have long texts. Longformer and reformer are models that try to be more efficient and 
+Most transformer models use full attention in the sense that the attention matrix is square. It can be a big
+computational bottleneck when you have long texts. Longformer and reformer are models that try to be more efficient and
 use a sparse version of the attention matrix to speed up training.
 
 .. _lsh-attention:
 
 **LSH attention**
 
-:ref:`Reformer <reformer>` uses LSH attention. In the softmax(QK^t), only the biggest elements (in the softmax 
-dimension) of the matrix QK^t are going to give useful contributions. So for each query q in Q, we can only consider 
-the keys k in K that are close to q. A hash function is used to determine if q and k are close. The attention mask is 
-modified to mask the current token (except at the first position) because it will give a query and key equal (so very 
-similar to each other). Since the hash can be a bit random, several hash functions are used in practice (determined by 
-a n_rounds parameter) then are averaged together.
+:ref:`Reformer <reformer>` uses LSH attention. In the softmax(QK^t), only the biggest elements (in the softmax
+dimension) of the matrix QK^t are going to give useful contributions. So for each query q in Q, we can  consider only
+the keys k in K that are close to q. A hash function is used to determine if q and k are close. The attention mask is
+modified to mask the current token (except at the first position), because it will give a query and a key equal (so very
+similar to each other). Since the hash can be a bit random, several hash functions are used in practice (determined by
+a n_rounds parameter) and then are averaged together.
 
 .. _local-attention:
 
 **Local attention**
 
-:ref:`Longformer <longformer>` uses local attention: often, the local context (e.g., what are the two tokens left and 
-right?) is enough to take action for a given token. Also, by stacking attention layers that have a small window, the 
-last layer will have a receptive field of more than just the tokens on the window, allowing them to build a 
+:ref:`Longformer <longformer>` uses local attention: often, the local context (e.g., what are the two tokens to the left and
+right?) is enough to take action for a given token. Also, by stacking attention layers that have a small window, the
+last layer will have a receptive field of more than just the tokens in the window, allowing them to build a
 representation of the whole sentence.
 
-Some preselected input tokens are also given global attention: for those few tokens, the attention matrix can access 
-all tokens and this process is symmetric: all other tokens have access to those specific tokens (on top of the ones in 
+Some preselected input tokens are also given global attention: for those few tokens, the attention matrix can access
+all tokens and this process is symmetric: all other tokens have access to those specific tokens (on top of the ones in
 their local window). This is shown in Figure 2d of the paper, see below for a sample attention mask:
 
 .. image:: imgs/local_attention_mask.png
    :scale: 50 %
    :align: center
 
-Using those attention matrices with less parameters then allows the model to have inputs having a bigger sequence 
+Using those attention matrices with less parameters then allows the model to have inputs having a bigger sequence
 length.
 
 Other tricks
@@ -606,13 +688,10 @@ Other tricks
 
 **Axial positional encodings**
 
-:ref:`Reformer <reformer>` uses axial positional encodings: in traditional transformer models, the positional encoding 
-E is a matrix of size :math:`l` by :math:`d`, :math:`l` being the sequence length and :math:`d` the dimension of the 
-hidden state. If you have very long texts, this matrix can be huge and take way too much space on the GPU.
-
-To alleviate that, axial positional encodings consists in factorizing that big matrix E in two smaller matrices E1 and 
+:ref:`Reformer <reformer>` uses axial positional encodings: in traditional transformer models, the positional encoding
+E is a matrix of size :math:`l` by :math:`d`, :math:`l` being the sequence length and :math:`d` the dimension of the
+hidden state. If you have very long texts, this matrix can be huge and take way too much space on the GPU. To alleviate that, axial positional encodings consist of factorizing that big matrix E in two smaller matrices E1 and
 E2, with dimensions :math:`l_{1} \times d_{1}` and :math:`l_{2} \times d_{2}`, such that :math:`l_{1} \times l_{2} = l`
-and :math:`d_{1} + d_{2} = d` (with the product for the lengths, this ends up being way smaller). The embedding for 
-time step :math:`j` in E is obtained by concatenating the embeddings for timestep :math:`j \% l1` in E1 and 
+and :math:`d_{1} + d_{2} = d` (with the product for the lengths, this ends up being way smaller). The embedding for
+time step :math:`j` in E is obtained by concatenating the embeddings for timestep :math:`j \% l1` in E1 and
 :math:`j // l1` in E2.
-

docs/source/perplexity.rst

@@ -0,0 +1,151 @@
+Perplexity of fixed-length models
+=================================
+
+Perplexity (PPL) is one of the most common metrics for evaluating language
+models. Before diving in, we should note that the metric applies specifically
+to classical language models (sometimes called autoregressive or causal
+language models) and is not well defined for masked language models like BERT
+(see :doc:`summary of the models <model_summary>`).
+
+Perplexity is defined as the exponentiated average log-likelihood of a
+sequence. If we have a tokenized sequence :math:`X = (x_0, x_1, \dots, x_t)`,
+then the perplexity of :math:`X` is,
+
+.. math::
+
+    \text{PPL}(X)
+    = \exp \left\{ {-\frac{1}{t}\sum_i^t \log p_\theta (x_i|x_{<i}) } \right\}
+
+where :math:`\log p_\theta (x_i|x_{<i})` is the log-likelihood of the ith
+token conditioned on the preceding tokens :math:`x_{<i}` according to our
+model. Intuitively, it can be thought of as an evaluation of the model's
+ability to predict uniformly among the set of specified tokens in a corpus.
+Importantly, this means that the tokenization procedure has a direct impact
+on a model's perplexity which should always be taken into consideration when
+comparing different models.
+
+This is also equivalent to the exponentiation of the cross-entropy between
+the data and model predictions. For more intuition about perplexity and its
+relationship to Bits Per Character (BPC) and data compression, check out this
+`fantastic blog post on The Gradient
+<https://thegradient.pub/understanding-evaluation-metrics-for-language-models/>`_.
+
+Calculating PPL with fixed-length models
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+If we weren't limited by a model's context size, we would evaluate the
+model's perplexity by autoregressively factorizing a sequence and
+conditioning on the entire preceding subsequence at each step, as shown
+below.
+
+.. image:: imgs/ppl_full.gif
+    :width: 600
+    :alt: Full decomposition of a sequence with unlimited context length
+
+When working with approximate models, however, we typically have a constraint
+on the number of tokens the model can process. The largest version
+of :doc:`GPT-2 <model_doc/gpt2>`, for example, has a fixed length of 1024
+tokens, so we cannot calculate :math:`p_\theta(x_t|x_{<t})` directly when
+:math:`t` is greater than 1024.
+
+Instead, the sequence is typically broken into subsequences equal to the
+model's maximum input size. If a model's max input size is :math:`k`, we
+then approximate the likelihood of a token :math:`x_t` by conditioning only
+on the :math:`k-1` tokens that precede it rather than the entire context.
+When evaluating the model's perplexity of a sequence, a tempting but
+suboptimal approach is to break the sequence into disjoint chunks and
+add up the decomposed log-likelihoods of each segment independently.
+
+.. image:: imgs/ppl_chunked.gif
+    :width: 600
+    :alt: Suboptimal PPL not taking advantage of full available context
+
+This is quick to compute since the perplexity of each segment can be computed
+in one forward pass, but serves as a poor approximation of the
+fully-factorized perplexity and will typically yield a higher (worse) PPL
+because the model will have less context at most of the prediction steps.
+
+Instead, the PPL of fixed-length models should be evaluated with a
+sliding-window strategy. This involves repeatedly sliding the
+context window so that the model has more context when making each
+prediction.
+
+.. image:: imgs/ppl_sliding.gif
+    :width: 600
+    :alt: Sliding window PPL taking advantage of all available context
+
+This is a closer approximation to the true decomposition of the
+sequence probability and will typically yield a more favorable score.
+The downside is that it requires a separate forward pass for each token in
+the corpus. A good practical compromise is to employ a strided sliding
+window, moving the context by larger strides rather than sliding by 1 token a
+time. This allows computation to procede much faster while still giving the
+model a large context to make predictions at each step.
+
+Example: Calculating perplexity with GPT-2 in 🤗 Transformers
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Let's demonstrate this process with GPT-2.
+
+.. code-block:: python
+
+    from transformers import GPT2LMHeadModel, GPT2TokenizerFast
+    device = 'cuda'
+    model_id = 'gpt2-large'
+    model = GPT2LMHeadModel.from_pretrained(model_id).to(device)
+    tokenizer = GPT2TokenizerFast.from_pretrained(model_id)
+
+We'll load in the WikiText-2 dataset and evaluate the perplexity using a few
+different sliding-window strategies. Since this dataset is small and we're
+just doing one forward pass over the set, we can just load and encode the
+entire dataset in memory.
+
+.. code-block:: python
+
+    from nlp import load_dataset
+    test = load_dataset('wikitext', 'wikitext-2-raw-v1', split='test')
+    encodings = tokenizer('\n\n'.join(test['text']), return_tensors='pt')
+
+With 🤗 Transformers, we can simply pass the ``input_ids`` as the ``labels``
+to our model, and the average log-likelihood for each token is returned as
+the loss. With our sliding window approach, however, there is overlap in the
+tokens we pass to the model at each iteration. We don't want the
+log-likelihood for the tokens we're just treating as context to be included
+in our loss, so we can set these targets to ``-100`` so that they are
+ignored. The following is an example of how we could do this with a stride of
+``512``. This means that the model will have at least 512 tokens for context
+when calculating the conditional likelihood of any one token (provided there
+are 512 preceding tokens available to condition on).
+
+.. code-block:: python
+
+    max_length = model.config.n_positions
+    stride = 512
+
+    lls = []
+    for i in tqdm(range(0, encodings.input_ids.size(1), stride)):
+        begin_loc = max(i + stride - max_length, 0)
+        end_loc = i + stride
+        input_ids = encodings.input_ids[:,begin_loc:end_loc].to(device)
+        target_ids = input_ids.clone()
+        target_ids[:,:-stride] = -100
+
+        with torch.no_grad():
+            outputs = model(input_ids, labels=target_ids)
+            log_likelihood = outputs[0] * stride
+
+        lls.append(log_likelihood)
+    
+    ppl = torch.exp(torch.stack(lls).sum() / i)
+
+Running this with the stride length equal to the max input length is
+equivalent to the suboptimal, non-sliding-window strategy we discussed above.
+The smaller the stride, the more context the model will have in making each
+prediction, and the better the reported perplexity will typically be.
+
+When we run the above with ``stride = 1024``, i.e. no overlap, the resulting
+PPL is ``19.64``, which is about the same as the ``19.93`` reported in the
+GPT-2 paper. By using ``stride = 512`` and thereby employing our striding
+window strategy, this jumps down to ``16.53``. This is not only a more
+favorable score, but is calculated in a way that is closer to the true
+autoregressive decomposition of a sequence likelihood.

docs/source/philosophy.rst

@@ -45,12 +45,12 @@ A few other goals:
     - A simple/consistent way to add new tokens to the vocabulary and embeddings for fine-tuning.
     - Simple ways to mask and prune transformer heads.
 
-- Switch easily between PyTorch and TensorFlow 2.0, allowing training using one framwork and inference using another.
+- Switch easily between PyTorch and TensorFlow 2.0, allowing training using one framework and inference using another.
 
 Main concepts
 ~~~~~~~~~~~~~
 
-The library is build around three types of classes for each model:
+The library is built around three types of classes for each model:
 
 - **Model classes**  such as :class:`~transformers.BertModel`, which are 30+ PyTorch models 
   (`torch.nn.Module <https://pytorch.org/docs/stable/nn.html#torch.nn.Module>`__) or Keras models 
@@ -65,9 +65,9 @@ The library is build around three types of classes for each model:
 
 All these classes can be instantiated from pretrained instances and saved locally using two methods:
 
-- :obj:`from_pretrained()` let you instantiate a model/configuration/tokenizer from a pretrained version either
+- :obj:`from_pretrained()` lets you instantiate a model/configuration/tokenizer from a pretrained version either
   provided by the library itself (the suported models are provided in the list :doc:`here <pretrained_models>`
   or stored locally (or on a server) by the user,
-- :obj:`save_pretrained()` let you save a model/configuration/tokenizer locally so that it can be reloaded using
+- :obj:`save_pretrained()` lets you save a model/configuration/tokenizer locally so that it can be reloaded using
   :obj:`from_pretrained()`.
 

docs/source/preprocessing.rst

@@ -20,7 +20,7 @@ work properly.
 To automatically download the vocab used during pretraining or fine-tuning a given model, you can use the 
 :func:`~transformers.AutoTokenizer.from_pretrained` method:
 
-::
+.. code-block::
 
     from transformers import AutoTokenizer
     tokenizer = AutoTokenizer.from_pretrained('bert-base-cased')
@@ -31,56 +31,41 @@ Base use
 A :class:`~transformers.PreTrainedTokenizer` has many methods, but the only one you need to remember for preprocessing
 is its ``__call__``: you just need to feed your sentence to your tokenizer object.
 
-::
-
-    encoded_input = tokenizer("Hello, I'm a single sentence!")
-    print(encoded_input)
-
-This will return a dictionary string to list of ints like this one:
-
-::
+.. code-block::
 
+    >>> encoded_input = tokenizer("Hello, I'm a single sentence!")
+    >>> print(encoded_input)
     {'input_ids': [101, 138, 18696, 155, 1942, 3190, 1144, 1572, 13745, 1104, 159, 9664, 2107, 102], 
      'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 
      'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}
 
+This returns a dictionary string to list of ints.
 The `input_ids <glossary.html#input-ids>`__ are the indices corresponding to each token in our sentence. We will see
 below what the `attention_mask <glossary.html#attention-mask>`__ is used for and in
 :ref:`the next section <sentence-pairs>` the goal of `token_type_ids <glossary.html#token-type-ids>`__.
 
 The tokenizer can decode a list of token ids in a proper sentence:
 
-::
-
-    tokenizer.decode(encoded_input["input_ids"])
-
-which should return
-
-::
+.. code-block::
 
+    >>> tokenizer.decode(encoded_input["input_ids"])
     "[CLS] Hello, I'm a single sentence! [SEP]"
 
 As you can see, the tokenizer automatically added some special tokens that the model expect. Not all model need special
 tokens; for instance, if we had used` gtp2-medium` instead of `bert-base-cased` to create our tokenizer, we would have
-
 seen the same sentence as the original one here. You can disable this behavior (which is only advised if you have added
 those special tokens yourself) by passing ``add_special_tokens=False``.
 
 If you have several sentences you want to process, you can do this efficiently by sending them as a list to the
 tokenizer:
 
-::
-
-    batch_sentences = ["Hello I'm a single sentence",
-                       "And another sentence",
-                       "And the very very last one"]
-    encoded_inputs = tokenizer(batch_sentences)
-    print(encoded_inputs)
-
-We get back a dictionary once again, this time with values being list of list of ints:
-
-::
+.. code-block::
 
+    >>> batch_sentences = ["Hello I'm a single sentence",
+    ...                    "And another sentence",
+    ...                    "And the very very last one"]
+    >>> encoded_inputs = tokenizer(batch_sentences)
+    >>> print(encoded_inputs)
     {'input_ids': [[101, 8667, 146, 112, 182, 170, 1423, 5650, 102],
                    [101, 1262, 1330, 5650, 102],
                    [101, 1262, 1103, 1304, 1304, 1314, 1141, 102]],
@@ -91,6 +76,8 @@ We get back a dictionary once again, this time with values being list of list of
                         [1, 1, 1, 1, 1],
                         [1, 1, 1, 1, 1, 1, 1, 1]]}
 
+We get back a dictionary once again, this time with values being list of list of ints.
+
 If the purpose of sending several sentences at a time to the tokenizer is to build a batch to feed the model, you will
 probably want:
 
@@ -100,19 +87,11 @@ probably want:
 
 You can do all of this by using the following options when feeding your list of sentences to the tokenizer:
 
-::
-
-    ## PYTORCH CODE
-    batch = tokenizer(batch_sentences, padding=True, truncation=True, return_tensors="pt")
-    print(batch)
-    ## TENSORFLOW CODE
-    batch = tokenizer(batch_sentences, padding=True, truncation=True, return_tensors="tf")
-    print(batch)
-
-which should now return a dictionary string to tensor like this:
-
-::
+.. code-block::
 
+    >>> ## PYTORCH CODE
+    >>> batch = tokenizer(batch_sentences, padding=True, truncation=True, return_tensors="pt")
+    >>> print(batch)
     {'input_ids': tensor([[ 101, 8667,  146,  112,  182,  170, 1423, 5650,  102],
                           [ 101, 1262, 1330, 5650,  102,    0,    0,    0,    0],
                           [ 101, 1262, 1103, 1304, 1304, 1314, 1141,  102,    0]]),
@@ -122,9 +101,22 @@ which should now return a dictionary string to tensor like this:
      'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1],
                                [1, 1, 1, 1, 1, 0, 0, 0, 0],
                                [1, 1, 1, 1, 1, 1, 1, 1, 0]])}
+    >>> ## TENSORFLOW CODE
+    >>> batch = tokenizer(batch_sentences, padding=True, truncation=True, return_tensors="tf")
+    >>> print(batch)
+    {'input_ids': tf.Tensor([[ 101, 8667,  146,  112,  182,  170, 1423, 5650,  102],
+                          [ 101, 1262, 1330, 5650,  102,    0,    0,    0,    0],
+                          [ 101, 1262, 1103, 1304, 1304, 1314, 1141,  102,    0]]),
+     'token_type_ids': tf.Tensor([[0, 0, 0, 0, 0, 0, 0, 0, 0],
+                               [0, 0, 0, 0, 0, 0, 0, 0, 0],
+                               [0, 0, 0, 0, 0, 0, 0, 0, 0]]), 
+     'attention_mask': tf.Tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1],
+                               [1, 1, 1, 1, 1, 0, 0, 0, 0],
+                               [1, 1, 1, 1, 1, 1, 1, 1, 0]])}
 
-We can now see what the `attention_mask <glossary.html#attention-mask>`__ is all about: it points out which tokens the
-model should pay attention to and which ones it should not (because they represent padding in this case).
+It returns a dictionary string to tensor. We can now see what the `attention_mask <glossary.html#attention-mask>`__ is
+all about: it points out which tokens the model should pay attention to and which ones it should not (because they
+represent padding in this case).
 
 
 Note that if your model does not have a maximum length associated to it, the command above will throw a warning. You
@@ -137,26 +129,16 @@ Preprocessing pairs of sentences
 
 Sometimes you need to feed pair of sentences to your model. For instance, if you want to classify if two sentences in a
 pair are similar, or for question-answering models, which take a context and a question. For BERT models, the input is
-then represented like this:
-
-::
-
-    [CLS] Sequence A [SEP] Sequence B [SEP]
+then represented like this: :obj:`[CLS] Sequence A [SEP] Sequence B [SEP]`
 
 You can encode a pair of sentences in the format expected by your model by supplying the two sentences as two arguments
-
 (not a list since a list of two sentences will be interpreted as a batch of two single sentences, as we saw before).
-
-
-::
-
-    encoded_input = tokenizer("How old are you?", "I'm 6 years old")
-    print(encoded_input)
-
 This will once again return a dict string to list of ints:
 
-::
+.. code-block::
 
+    >>> encoded_input = tokenizer("How old are you?", "I'm 6 years old")
+    >>> print(encoded_input)
     {'input_ids': [101, 1731, 1385, 1132, 1128, 136, 102, 146, 112, 182, 127, 1201, 1385, 102], 
      'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1], 
      'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}
@@ -169,34 +151,24 @@ using ``return_input_ids`` or ``return_token_type_ids``.
 
 If we decode the token ids we obtained, we will see that the special tokens have been properly added.
 
-::
-
-    tokenizer.decode(encoded_input["input_ids"])
-
-will return:
-
-::
+.. code-block::
 
+    >>> tokenizer.decode(encoded_input["input_ids"])
     "[CLS] How old are you? [SEP] I'm 6 years old [SEP]"
 
 If you have a list of pairs of sequences you want to process, you should feed them as two lists to your tokenizer: the
 list of first sentences and the list of second sentences:
 
-::
-
-    batch_sentences = ["Hello I'm a single sentence",
-                       "And another sentence",
-                       "And the very very last one"]
-    batch_of_second_sentences = ["I'm a sentence that goes with the first sentence",
-                                 "And I should be encoded with the second sentence",
-                                 "And I go with the very last one"]
-    encoded_inputs = tokenizer(batch_sentences, batch_of_second_sentences)
-    print(encoded_inputs)
-
-will return a dict with the values being list of lists of ints:
-
-::
+.. code-block::
 
+    >>> batch_sentences = ["Hello I'm a single sentence",
+    ...                    "And another sentence",
+    ...                    "And the very very last one"]
+    >>> batch_of_second_sentences = ["I'm a sentence that goes with the first sentence",
+    ...                              "And I should be encoded with the second sentence",
+    ...                              "And I go with the very last one"]
+    >>> encoded_inputs = tokenizer(batch_sentences, batch_of_second_sentences)
+    >>> print(encoded_inputs)
     {'input_ids': [[101, 8667, 146, 112, 182, 170, 1423, 5650, 102, 146, 112, 182, 170, 5650, 1115, 2947, 1114, 1103, 1148, 5650, 102], 
                    [101, 1262, 1330, 5650, 102, 1262, 146, 1431, 1129, 12544, 1114, 1103, 1248, 5650, 102], 
                    [101, 1262, 1103, 1304, 1304, 1314, 1141, 102, 1262, 146, 1301, 1114, 1103, 1304, 1314, 1141, 102]], 
@@ -207,17 +179,14 @@ will return a dict with the values being list of lists of ints:
                        [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], 
                        [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]}
 
+As we can see, it returns a dictionary with the values being list of lists of ints.
+
 To double-check what is fed to the model, we can decode each list in `input_ids` one by one:
 
-::
-
-    for ids in encoded_inputs["input_ids"]:
-        print(tokenizer.decode(ids))
-
-which will return:
-
-::
+.. code-block::
 
+    >>> for ids in encoded_inputs["input_ids"]:
+    >>>     print(tokenizer.decode(ids))
     [CLS] Hello I'm a single sentence [SEP] I'm a sentence that goes with the first sentence [SEP]
     [CLS] And another sentence [SEP] And I should be encoded with the second sentence [SEP]
     [CLS] And the very very last one [SEP] And I go with the very last one [SEP]
@@ -225,7 +194,7 @@ which will return:
 Once again, you can automatically pad your inputs to the maximum sentence length in the batch, truncate to the maximum
 length the model can accept and return tensors directly with the following:
 
-::
+.. code-block::
 
     ## PYTORCH CODE
     batch = tokenizer(batch_sentences, batch_of_second_sentences, padding=True, truncation=True, return_tensors="pt")
@@ -315,18 +284,19 @@ The tokenizer also accept pre-tokenized inputs. This is particularly useful when
 predictions in `named entity recognition (NER) <https://en.wikipedia.org/wiki/Named-entity_recognition>`__ or
 `part-of-speech tagging (POS tagging) <https://en.wikipedia.org/wiki/Part-of-speech_tagging>`__.
 
+.. warning::
+
+    Pre-tokenized does not mean your inputs are already tokenized (you wouldn't need to pass them though the tokenizer
+    if that was the case) but just split into words (which is often the first step in subword tokenization algorithms
+    like BPE).
+
 If you want to use pre-tokenized inputs, just set :obj:`is_pretokenized=True` when passing your inputs to the
-tokenizer. For instance:
+tokenizer. For instance, we have:
 
-::
-
-    encoded_input = tokenizer(["Hello", "I'm", "a", "single", "sentence"], is_pretokenized=True)
-    print(encoded_input)
-
-will return:
-
-::
+.. code-block::
 
+    >>> encoded_input = tokenizer(["Hello", "I'm", "a", "single", "sentence"], is_pretokenized=True)
+    >>> print(encoded_input)
     {'input_ids': [101, 8667, 146, 112, 182, 170, 1423, 5650, 102],
      'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0], 
      'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1]}
@@ -337,7 +307,7 @@ Note that the tokenizer still adds the ids of special tokens (if applicable) unl
 This works exactly as before for batch of sentences or batch of pairs of sentences. You can encode a batch of sentences
 like this:
 
-::
+.. code-block::
 
     batch_sentences = [["Hello", "I'm", "a", "single", "sentence"],
                        ["And", "another", "sentence"],
@@ -346,7 +316,7 @@ like this:
 
 or a batch of pair sentences like this:
 
-::
+.. code-block::
 
     batch_of_second_sentences = [["I'm", "a", "sentence", "that", "goes", "with", "the", "first", "sentence"],
                                  ["And", "I", "should", "be", "encoded", "with", "the", "second", "sentence"],
@@ -355,7 +325,7 @@ or a batch of pair sentences like this:
 
 And you can add padding, truncation as well as directly return tensors like before:
 
-::
+.. code-block::
 
     ## PYTORCH CODE
     batch = tokenizer(batch_sentences,

docs/source/pretrained_models.rst

@@ -74,14 +74,16 @@ For a list that includes community-uploaded models, refer to `https://huggingfac
 |                   |                                                            | (see `details on dbmdz repository <https://github.com/dbmdz/german-bert>`__).                                                         |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``cl-tohoku/bert-base-japanese``                           | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
-|                   |                                                            | | Trained on Japanese text. Text is tokenized with MeCab and WordPiece.                                                               |
-|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                          |
+|                   |                                                            | | Trained on Japanese text. Text is tokenized with MeCab and WordPiece and this requires some extra dependencies,                     |
+|                   |                                                            | | `fugashi <https://github.com/polm/fugashi>`__ which is a wrapper around `MeCab <https://taku910.github.io/mecab/>`__.               |
+|                   |                                                            | | Use ``pip install transformers["ja"]`` (or ``pip install -e .["ja"]`` if you install from source) to install them.                  |
 |                   |                                                            |                                                                                                                                       |
 |                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``cl-tohoku/bert-base-japanese-whole-word-masking``        | | 12-layer, 768-hidden, 12-heads, 110M parameters.                                                                                    |
-|                   |                                                            | | Trained on Japanese text using Whole-Word-Masking. Text is tokenized with MeCab and WordPiece.                                      |
-|                   |                                                            | | `MeCab <https://taku910.github.io/mecab/>`__ is required for tokenization.                                                          |
+|                   |                                                            | | Trained on Japanese text. Text is tokenized with MeCab and WordPiece and this requires some extra dependencies,                     |
+|                   |                                                            | | `fugashi <https://github.com/polm/fugashi>`__ which is a wrapper around `MeCab <https://taku910.github.io/mecab/>`__.               |
+|                   |                                                            | | Use ``pip install transformers["ja"]`` (or ``pip install -e .["ja"]`` if you install from source) to install them.                  |
 |                   |                                                            |                                                                                                                                       |
 |                   |                                                            | (see `details on cl-tohoku repository <https://github.com/cl-tohoku/bert-japanese>`__).                                               |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
@@ -329,9 +331,6 @@ For a list that includes community-uploaded models, refer to `https://huggingfac
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``facebook/bart-large-cnn``                                | | 12-layer, 1024-hidden, 16-heads, 406M parameters       (same as base)                                                               |
 |                   |                                                            | | bart-large base architecture finetuned on cnn summarization task                                                                    |
-|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
-|                   | ``facebook/mbart-large-en-ro``                             | | 12-layer, 1024-hidden, 16-heads, 880M parameters                                                                                    |
-|                   |                                                            | | bart-large architecture pretrained on cc25 multilingual data , finetuned on WMT english romanian translation.                       |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | DialoGPT          | ``DialoGPT-small``                                         | | 12-layer, 768-hidden, 12-heads, 124M parameters                                                                                     |
 |                   |                                                            | | Trained on English text: 147M conversation-like exchanges extracted from Reddit.                                                    |
@@ -351,9 +350,17 @@ For a list that includes community-uploaded models, refer to `https://huggingfac
 | MarianMT          | ``Helsinki-NLP/opus-mt-{src}-{tgt}``                       | | 12-layer, 512-hidden, 8-heads, ~74M parameter Machine translation models. Parameter counts vary depending on vocab size.            |
 |                   |                                                            | | (see `model list <https://huggingface.co/Helsinki-NLP>`_)                                                                           |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+| Pegasus           | ``google/pegasus-{dataset}``                               | | 16-layer, 1024-hidden, 16-heads, ~568M parameter, 2.2 GB for summary. `model list <https://huggingface.co/models?search=pegasus>`__ |
++-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 | Longformer        | ``allenai/longformer-base-4096``                           | | 12-layer, 768-hidden, 12-heads, ~149M parameters                                                                                    |
 |                   |                                                            | | Starting from RoBERTa-base checkpoint, trained on documents of max length 4,096                                                     |
 |                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
 |                   | ``allenai/longformer-large-4096``                          | | 24-layer, 1024-hidden, 16-heads, ~435M parameters                                                                                   |
 |                   |                                                            | | Starting from RoBERTa-large checkpoint, trained on documents of max length 4,096                                                    |
 +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+| MBart             | ``facebook/mbart-large-cc25``                              | | 24-layer, 1024-hidden, 16-heads, 610M parameters                                                                                    |
+|                   |                                                            | | mBART (bart-large architecture) model trained on 25 languages' monolingual corpus                                                   |
+|                   +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
+|                   | ``facebook/mbart-large-en-ro``                             | | 24-layer, 1024-hidden, 16-heads, 610M parameters                                                                                    |
+|                   |                                                            | | mbart-large-cc25 model finetuned on WMT english romanian translation.                                                               |
++-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+

docs/source/quicktour.rst

@@ -108,11 +108,11 @@ any other model from the model hub):
     >>> model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
     >>> model = AutoModelForSequenceClassification.from_pretrained(model_name)
     >>> tokenizer = AutoTokenizer.from_pretrained(model_name)
-    >>> pipe = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
+    >>> classifier = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
     >>> ## TENSORFLOW CODE
     >>> model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
     >>> # This model only exists in PyTorch, so we use the `from_pt` flag to import that model in TensorFlow.
-    >>> model = TFAutoModelForSequenceClassification.from_pretrained(model_name, from_pt=True) 
+    >>> model = TFAutoModelForSequenceClassification.from_pretrained(model_name, from_pt=True)
     >>> tokenizer = AutoTokenizer.from_pretrained(model_name)
     >>> classifier = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
 
@@ -128,7 +128,7 @@ Under the hood: pretrained models
 Let's now see what happens beneath the hood when using those pipelines. As we saw, the model and tokenizer are created
 using the :obj:`from_pretrained` method:
 
-::
+.. code-block::
 
     >>> ## PYTORCH CODE
     >>> from transformers import AutoTokenizer, AutoModelForSequenceClassification
@@ -146,8 +146,9 @@ Using the tokenizer
 
 We mentioned the tokenizer is responsible for the preprocessing of your texts. First, it will split a given text in
 words (or part of words, punctuation symbols, etc.) usually called `tokens`. There are multiple rules that can govern
-that process, which is why we need to instantiate the tokenizer using the name of the model, to make sure we use the
-same rules as when the model was pretrained.
+that process (you can learn more about them in the :doc:`tokenizer summary <tokenizer_summary>`, which is why we need
+to instantiate the tokenizer using the name of the model, to make sure we use the same rules as when the model was
+pretrained.
 
 The second step is to convert those `tokens` into numbers, to be able to build a tensor out of them and feed them to
 the model. To do this, the tokenizer has a `vocab`, which is the part we download when we instantiate it with the
@@ -190,7 +191,7 @@ and get tensors back. You can specify all of that to the tokenizer:
     ...     return_tensors="tf"
     ... )
 
-The padding is automatically applied on the side the model expect it (in this case, on the right), with the
+The padding is automatically applied on the side expected by the model (in this case, on the right), with the
 padding token the model was pretrained with. The attention mask is also adapted to take the padding into account:
 
 .. code-block::
@@ -211,9 +212,9 @@ You can learn more about tokenizers :doc:`here <preprocessing>`.
 Using the model
 ^^^^^^^^^^^^^^^
 
-Once your input has been preprocessed by the tokenizer, you can directly send it to the model. As we mentioned, it will
-contain all the relevant information the model needs. If you're using a TensorFlow model, you can directly pass the
-dictionary keys to tensor, for a PyTorch model, you need to unpack the dictionary by adding :obj:`**`.
+Once your input has been preprocessed by the tokenizer, you can send it directly to the model. As we mentioned, it will
+contain all the relevant information the model needs. If you're using a TensorFlow model, you can pass the
+dictionary keys directly to tensor, for a PyTorch model, you need to unpack the dictionary by adding :obj:`**`.
 
 .. code-block::
 
@@ -234,9 +235,11 @@ final activations of the model.
     >>> ## TENSORFLOW CODE
     >>> print(tf_outputs)
     (<tf.Tensor: shape=(2, 2), dtype=float32, numpy=
-    array([[-4.0832963 ,  4.3364134 ],
-           [ 0.08181238, -0.04178794]], dtype=float32)>,)
+    array([[-4.0832963 ,  4.336414  ],
+           [ 0.08181786, -0.04179301]], dtype=float32)>,)
 
+The model can return more than just the final activations, which is why the output is a tuple. Here we only asked for
+the final activations, so we get a tuple with one element.
 .. note::
 
     All 🤗 Transformers models (PyTorch or TensorFlow) return the activations of the model *before* the final
@@ -261,7 +264,7 @@ We can see we get the numbers from before:
     >>> print(tf_predictions)
     tf.Tensor(
     [[2.2042994e-04 9.9977952e-01]
-     [5.3086078e-01 4.6913919e-01]], shape=(2, 2), dtype=float32)
+     [5.3086340e-01 4.6913657e-01]], shape=(2, 2), dtype=float32)
     >>> ## PYTORCH CODE
     >>> print(pt_predictions)
     tensor([[2.2043e-04, 9.9978e-01],
@@ -284,9 +287,15 @@ training loop. 🤗 Transformers also provides a :class:`~transformers.Trainer`
 you are using TensorFlow) class to help with your training (taking care of things such as distributed training, mixed
 precision, etc.). See the :doc:`training tutorial <training>` for more details.
 
-Once your model is fine-tuned, you can save it with its tokenizer the following way:
+.. note::
 
-::
+    Pytorch model outputs are special dataclasses so that you can get autocompletion for their attributes in an IDE.
+    They also behave like a tuple or a dictionary (e.g., you can index with an integer, a slice or a string) in which
+    case the attributes not set (that have :obj:`None` values) are ignored.
+
+Once your model is fine-tuned, you can save it with its tokenizer in the following way:
+
+.. code-block::
 
     tokenizer.save_pretrained(save_directory)
     model.save_pretrained(save_directory)
@@ -296,14 +305,14 @@ directory name instead of the model name. One cool feature of 🤗 Transformers
 PyTorch and TensorFlow: any model saved as before can be loaded back either in PyTorch or TensorFlow. If you are
 loading a saved PyTorch model in a TensorFlow model, use :func:`~transformers.TFAutoModel.from_pretrained` like this:
 
-::
+.. code-block::
 
     tokenizer = AutoTokenizer.from_pretrained(save_directory)
     model = TFAutoModel.from_pretrained(save_directory, from_pt=True)
 
 and if you are loading a saved TensorFlow model in a PyTorch model, you should use the following code:
 
-::
+.. code-block::
 
     tokenizer = AutoTokenizer.from_pretrained(save_directory)
     model = AutoModel.from_pretrained(save_directory, from_tf=True)
@@ -311,7 +320,7 @@ and if you are loading a saved TensorFlow model in a PyTorch model, you should u
 Lastly, you can also ask the model to return all hidden states and all attention weights if you need them:
 
 
-::
+.. code-block::
 
     >>> ## PYTORCH CODE
     >>> pt_outputs = pt_model(**pt_batch, output_hidden_states=True, output_attentions=True)
@@ -328,7 +337,9 @@ pretrained model. Behind the scenes, the library has one model class per combina
 code is easy to access and tweak if you need to.
 
 In our previous example, the model was called "distilbert-base-uncased-finetuned-sst-2-english", which means it's
-using the :doc:`DistilBERT </model_doc/distilbert>` architecture. The model automatically created is then a
+using the :doc:`DistilBERT </model_doc/distilbert>` architecture. As
+:class:`~transformers.AutoModelForSequenceClassification` (or :class:`~transformers.TFAutoModelForSequenceClassification`
+if you are using TensorFlow) was used, the model automatically created is then a
 :class:`~transformers.DistilBertForSequenceClassification`. You can look at its documentation for all details relevant
 to that specific model, or browse the source code. This is how you would directly instantiate model and tokenizer
 without the auto magic:
@@ -351,7 +362,7 @@ Customizing the model
 
 If you want to change how the model itself is built, you can define your custom configuration class. Each architecture
 comes with its own relevant configuration (in the case of DistilBERT, :class:`~transformers.DistilBertConfig`) which
-allows you to specify any of the hidden dimension, dropout rate etc. If you do core modifications, like changing the
+allows you to specify any of the hidden dimension, dropout rate, etc. If you do core modifications, like changing the
 hidden size, you won't be able to use a pretrained model anymore and will need to train from scratch. You would then
 instantiate the model directly from this configuration.
 

docs/source/serialization.rst

@@ -0,0 +1,251 @@
+**********************************************
+Exporting transformers models
+**********************************************
+
+ONNX / ONNXRuntime
+==============================================
+
+Projects `ONNX (Open Neural Network eXchange) <http://onnx.ai>`_ and `ONNXRuntime (ORT) <https://microsoft.github.io/onnxruntime/>`_ are part of an effort from leading industries in the AI field
+to provide a unified and community-driven format to store and, by extension, efficiently execute neural network leveraging a variety
+of hardware and dedicated optimizations.
+
+Starting from transformers v2.10.0 we partnered with ONNX Runtime to provide an easy export of transformers models to
+the ONNX format. You can have a look at the effort by looking at our joint blog post `Accelerate your NLP pipelines using
+Hugging Face Transformers and ONNX Runtime <https://medium.com/microsoftazure/accelerate-your-nlp-pipelines-using-hugging-face-transformers-and-onnx-runtime-2443578f4333>`_.
+
+Exporting a model is done through the script `convert_graph_to_onnx.py` at the root of the transformers sources.
+The following command shows how easy it is to export a BERT model from the library, simply run:
+
+.. code-block:: bash
+
+    python convert_graph_to_onnx.py --framework <pt, tf> --model bert-base-cased bert-base-cased.onnx
+
+The conversion tool works for both PyTorch and Tensorflow models and ensures:
+
+* The model and its weights are correctly initialized from the Hugging Face model hub or a local checkpoint.
+* The inputs and outputs are correctly generated to their ONNX counterpart.
+* The generated model can be correctly loaded through onnxruntime.
+
+.. note::
+    Currently, inputs and outputs are always exported with dynamic sequence axes preventing some optimizations
+    on the ONNX Runtime. If you would like to see such support for fixed-length inputs/outputs, please
+    open up an issue on transformers.
+
+
+Also, the conversion tool supports different options which let you tune the behavior of the generated model:
+
+* **Change the target opset version of the generated model.**  (More recent opset generally supports more operators and enables faster inference)
+
+* **Export pipeline-specific prediction heads.**  (Allow to export model along with its task-specific prediction head(s))
+
+* **Use the external data format (PyTorch only).**  (Lets you export model which size is above 2Gb (`More info <https://github.com/pytorch/pytorch/pull/33062>`_))
+
+
+Optimizations
+------------------------------------------------
+
+ONNXRuntime includes some transformers-specific transformations to leverage optimized operations in the graph.
+Below are some of the operators which can be enabled to speed up inference through ONNXRuntime (*see note below*):
+
+* Constant folding
+* Attention Layer fusing
+* Skip connection LayerNormalization fusing
+* FastGeLU approximation
+
+Some of the optimizations performed by ONNX runtime can be hardware specific and thus lead to different performances
+if used on another machine with a different hardware configuration than the one used for exporting the model.
+For this reason, when using ``convert_graph_to_onnx.py`` optimizations are not enabled,
+ensuring the model can be easily exported to various hardware.
+Optimizations can then be enabled when loading the model through ONNX runtime for inference.
+
+
+.. note::
+    When quantization is enabled (see below), ``convert_graph_to_onnx.py`` script will enable optimizations on the model
+    because quantization would modify the underlying graph making it impossible for ONNX runtime to do the optimizations
+    afterwards.
+
+.. note::
+    For more information about the optimizations enabled by ONNXRuntime, please have a look at the (`ONNXRuntime Github <https://github.com/microsoft/onnxruntime/tree/master/onnxruntime/python/tools/transformers>`_)
+
+Quantization
+------------------------------------------------
+
+ONNX exporter supports generating a quantized version of the model to allow efficient inference.
+
+Quantization works by converting the memory representation of the parameters in the neural network
+to a compact integer format. By default, weights of a neural network are stored as single-precision float (`float32`)
+which can express a wide-range of floating-point numbers with decent precision.
+These properties are especially interesting at training where you want fine-grained representation.
+
+On the other hand, after the training phase, it has been shown one can greatly reduce the range and the precision of `float32` numbers
+without changing the performances of the neural network.
+
+More technically, `float32` parameters are converted to a type requiring fewer bits to represent each number, thus reducing
+the overall size of the model. Here, we are enabling `float32` mapping to `int8` values (a non-floating, single byte, number representation)
+according to the following formula:
+
+.. math::
+    y_{float32} = scale * x_{int8} - zero\_point
+
+.. note::
+    The quantization process will infer the parameter `scale` and `zero_point` from the neural network parameters
+
+Leveraging tiny-integers has numerous advantages when it comes to inference:
+
+* Storing fewer bits instead of 32 bits for the `float32` reduces the size of the model and makes it load faster.
+* Integer operations execute a magnitude faster on modern hardware
+* Integer operations require less power to do the computations
+
+In order to convert a transformers model to ONNX IR with quantized weights you just need to specify ``--quantize``
+when using ``convert_graph_to_onnx.py``. Also, you can have a look at the ``quantize()`` utility-method in this
+same script file.
+
+Example of quantized BERT model export:
+
+.. code-block:: bash
+
+    python convert_graph_to_onnx.py --framework <pt, tf> --model bert-base-cased --quantize bert-base-cased.onnx
+
+.. note::
+    Quantization support requires ONNX Runtime >= 1.4.0
+
+.. note::
+    When exporting quantized model you will end up with two different ONNX files. The one specified at the end of the
+    above command will contain the original ONNX model storing `float32` weights.
+    The second one, with ``-quantized`` suffix, will hold the quantized parameters.
+
+
+TorchScript
+=======================================
+
+.. note::
+    This is the very beginning of our experiments with TorchScript and we are still exploring its capabilities
+    with variable-input-size models. It is a focus of interest to us and we will deepen our analysis in upcoming
+    releases, with more code examples, a more flexible implementation, and benchmarks comparing python-based codes
+    with compiled TorchScript.
+
+
+According to Pytorch's documentation: "TorchScript is a way to create serializable and optimizable models from PyTorch code".
+Pytorch's two modules `JIT and TRACE <https://pytorch.org/docs/stable/jit.html>`_ allow the developer to export
+their model to be re-used in other programs, such as efficiency-oriented C++ programs.
+
+We have provided an interface that allows the export of 🤗 Transformers models to TorchScript so that they can
+be reused in a different environment than a Pytorch-based python program. Here we explain how to export and use our models using TorchScript.
+
+Exporting a model requires two things:
+
+* a forward pass with dummy inputs.
+* model instantiation with the ``torchscript`` flag.
+
+These necessities imply several things developers should be careful about. These are detailed below.
+
+
+Implications
+------------------------------------------------
+
+TorchScript flag and tied weights
+------------------------------------------------
+This flag is necessary because most of the language models in this repository have tied weights between their
+``Embedding`` layer and their ``Decoding`` layer. TorchScript does not allow the export of models that have tied weights, therefore
+it is necessary to untie and clone the weights beforehand.
+
+This implies that models instantiated with the ``torchscript`` flag have their ``Embedding`` layer and ``Decoding`` layer
+separate, which means that they should not be trained down the line. Training would de-synchronize the two layers,
+leading to unexpected results.
+
+This is not the case for models that do not have a Language Model head, as those do not have tied weights. These models
+can be safely exported without the ``torchscript`` flag.
+
+Dummy inputs and standard lengths
+------------------------------------------------
+
+The dummy inputs are used to do a model forward pass. While the inputs' values are propagating through the layers,
+Pytorch keeps track of the different operations executed on each tensor. These recorded operations are then used
+to create the "trace" of the model.
+
+The trace is created relatively to the inputs' dimensions. It is therefore constrained by the dimensions of the dummy
+input, and will not work for any other sequence length or batch size. When trying with a different size, an error such
+as:
+
+``The expanded size of the tensor (3) must match the existing size (7) at non-singleton dimension 2``
+
+will be raised. It is therefore recommended to trace the model with a dummy input size at least as large as the largest
+input that will be fed to the model during inference. Padding can be performed to fill the missing values. As the model
+will have been traced with a large input size however, the dimensions of the different matrix will be large as well,
+resulting in more calculations.
+
+It is recommended to be careful of the total number of operations done on each input and to follow performance closely
+when exporting varying sequence-length models.
+
+Using TorchScript in Python
+-------------------------------------------------
+
+Below is an example, showing how to save, load models as well as how to use the trace for inference.
+
+Saving a model
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+This snippet shows how to use TorchScript to export a ``BertModel``. Here the ``BertModel`` is instantiated
+according to a ``BertConfig`` class and then saved to disk under the filename ``traced_bert.pt``
+
+.. code-block:: python
+
+    from transformers import BertModel, BertTokenizer, BertConfig
+    import torch
+
+    enc = BertTokenizer.from_pretrained("bert-base-uncased")
+
+    # Tokenizing input text
+    text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
+    tokenized_text = enc.tokenize(text)
+
+    # Masking one of the input tokens
+    masked_index = 8
+    tokenized_text[masked_index] = '[MASK]'
+    indexed_tokens = enc.convert_tokens_to_ids(tokenized_text)
+    segments_ids = [0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1]
+
+    # Creating a dummy input
+    tokens_tensor = torch.tensor([indexed_tokens])
+    segments_tensors = torch.tensor([segments_ids])
+    dummy_input = [tokens_tensor, segments_tensors]
+
+    # Initializing the model with the torchscript flag
+    # Flag set to True even though it is not necessary as this model does not have an LM Head.
+    config = BertConfig(vocab_size_or_config_json_file=32000, hidden_size=768,
+        num_hidden_layers=12, num_attention_heads=12, intermediate_size=3072, torchscript=True)
+
+    # Instantiating the model
+    model = BertModel(config)
+
+    # The model needs to be in evaluation mode
+    model.eval()
+
+    # If you are instantiating the model with `from_pretrained` you can also easily set the TorchScript flag
+    model = BertModel.from_pretrained("bert-base-uncased", torchscript=True)
+
+    # Creating the trace
+    traced_model = torch.jit.trace(model, [tokens_tensor, segments_tensors])
+    torch.jit.save(traced_model, "traced_bert.pt")
+
+Loading a model
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+This snippet shows how to load the ``BertModel`` that was previously saved to disk under the name ``traced_bert.pt``.
+We are re-using the previously initialised ``dummy_input``.
+
+.. code-block:: python
+
+    loaded_model = torch.jit.load("traced_bert.pt")
+    loaded_model.eval()
+
+    all_encoder_layers, pooled_output = loaded_model(*dummy_input)
+
+Using a traced model for inference
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Using the traced model for inference is as simple as using its ``__call__`` dunder method:
+
+.. code-block:: python
+
+    traced_model(tokens_tensor, segments_tensors)

docs/source/task_summary.rst

@@ -15,18 +15,17 @@ checkpoints are usually pre-trained on a large corpus of data and fine-tuned on
 following:
 
 - Not all models were fine-tuned on all tasks. If you want to fine-tune a model on a specific task, you can leverage
-  one of the `run_$TASK.py` script in the
-  `examples <https://github.com/huggingface/transformers/tree/master/examples>`_ directory.
+  one of the `run_$TASK.py` scripts in the
+  `examples <https://github.com/huggingface/transformers/tree/master/examples>`__ directory.
 - Fine-tuned models were fine-tuned on a specific dataset. This dataset may or may not overlap with your use-case
   and domain. As mentioned previously, you may leverage the
-  `examples <https://github.com/huggingface/transformers/tree/master/examples>`_ scripts to fine-tune your model, or you
+  `examples <https://github.com/huggingface/transformers/tree/master/examples>`__ scripts to fine-tune your model, or you
   may create your own training script.
 
 In order to do an inference on a task, several mechanisms are made available by the library:
 
 - Pipelines: very easy-to-use abstractions, which require as little as two lines of code.
-- Using a model directly with a tokenizer (PyTorch/TensorFlow): the full inference using the model. Less abstraction,
-  but much more powerful.
+- Direct model use: Less abstractions, but more flexibility and power via a direct access to a tokenizer (PyTorch/TensorFlow) and full inference capacity.
 
 Both approaches are showcased here.
 
@@ -44,10 +43,11 @@ Sequence Classification
 Sequence classification is the task of classifying sequences according to a given number of classes. An example
 of sequence classification is the GLUE dataset, which is entirely based on that task. If you would like to fine-tune
 a model on a GLUE sequence classification task, you may leverage the
-`run_glue.py <https://github.com/huggingface/transformers/tree/master/examples/text-classification/run_glue.py>`_ or
-`run_tf_glue.py <https://github.com/huggingface/transformers/tree/master/examples/text-classification/run_tf_glue.py>`_ scripts.
+`run_glue.py <https://github.com/huggingface/transformers/tree/master/examples/text-classification/run_glue.py>`__ and
+`run_pl_glue.py <https://github.com/huggingface/transformers/tree/master/examples/text-classification/run_pl_glue.py>`__ or
+`run_tf_glue.py <https://github.com/huggingface/transformers/tree/master/examples/text-classification/run_tf_glue.py>`__ scripts.
 
-Here is an example using the pipelines do to sentiment analysis: identifying if a sequence is positive or negative.
+Here is an example of using pipelines to do sentiment analysis: identifying if a sequence is positive or negative.
 It leverages a fine-tuned model on sst2, which is a GLUE task.
 
 This returns a label ("POSITIVE" or "NEGATIVE") alongside a score, as follows:
@@ -70,15 +70,17 @@ This returns a label ("POSITIVE" or "NEGATIVE") alongside a score, as follows:
 Here is an example of doing a sequence classification using a model to determine if two sequences are paraphrases
 of each other. The process is the following:
 
-- Instantiate a tokenizer and a model from the checkpoint name. The model is identified as a BERT model and loads it
-  with the weights stored in the checkpoint.
-- Build a sequence from the two sentences, with the correct model-specific separators token type ids
-  and attention masks (:func:`~transformers.PreTrainedTokenizer.encode` and
-  :func:`~transformers.PreTrainedTokenizer.__call__` take care of this)
-- Pass this sequence through the model so that it is classified in one of the two available classes: 0
-  (not a paraphrase) and 1 (is a paraphrase)
-- Compute the softmax of the result to get probabilities over the classes
-- Print the results
+1. Instantiate a tokenizer and a model from the checkpoint name. The model is
+   identified as a BERT model and loads it with the weights stored in the
+   checkpoint.
+2. Build a sequence from the two sentences, with the correct model-specific
+   separators token type ids and attention masks
+   (:func:`~transformers.PreTrainedTokenizer.encode` and
+   :func:`~transformers.PreTrainedTokenizer.__call__` take care of this).
+3. Pass this sequence through the model so that it is classified in one of the
+   two available classes: 0 (not a paraphrase) and 1 (is a paraphrase).
+4. Compute the softmax of the result to get probabilities over the classes.
+5. Print the results.
 
 .. code-block::
 
@@ -98,8 +100,8 @@ of each other. The process is the following:
     >>> paraphrase = tokenizer(sequence_0, sequence_2, return_tensors="pt")
     >>> not_paraphrase = tokenizer(sequence_0, sequence_1, return_tensors="pt")
 
-    >>> paraphrase_classification_logits = model(**paraphrase)[0]
-    >>> not_paraphrase_classification_logits = model(**not_paraphrase)[0]
+    >>> paraphrase_classification_logits = model(**paraphrase).logits
+    >>> not_paraphrase_classification_logits = model(**not_paraphrase).logits
 
     >>> paraphrase_results = torch.softmax(paraphrase_classification_logits, dim=1).tolist()[0]
     >>> not_paraphrase_results = torch.softmax(not_paraphrase_classification_logits, dim=1).tolist()[0]
@@ -154,9 +156,12 @@ Extractive Question Answering
 
 Extractive Question Answering is the task of extracting an answer from a text given a question. An example of a
 question answering dataset is the SQuAD dataset, which is entirely based on that task. If you would like to fine-tune
-a model on a SQuAD task, you may leverage the `run_squad.py`.
+a model on a SQuAD task, you may leverage the
+`run_squad.py <https://github.com/huggingface/transformers/tree/master/examples/question-answering/run_squad.py>`__ and
+`run_tf_squad.py <https://github.com/huggingface/transformers/tree/master/examples/question-answering/run_tf_squad.py>`__ scripts.
 
-Here is an example using the pipelines do to question answering: extracting an answer from a text given a question.
+
+Here is an example of using pipelines to do question answering: extracting an answer from a text given a question.
 It leverages a fine-tuned model on SQuAD.
 
 .. code-block::
@@ -171,7 +176,7 @@ It leverages a fine-tuned model on SQuAD.
     ... a model on a SQuAD task, you may leverage the examples/question-answering/run_squad.py script.
     ... """
 
-This returns an answer extracted from the text, a confidence score, alongside "start" and "end" values which
+This returns an answer extracted from the text, a confidence score, alongside "start" and "end" values, which
 are the positions of the extracted answer in the text.
 
 .. code-block::
@@ -187,16 +192,19 @@ are the positions of the extracted answer in the text.
 
 Here is an example of question answering using a model and a tokenizer. The process is the following:
 
-- Instantiate a tokenizer and a model from the checkpoint name. The model is identified as a BERT model and loads it
-  with the weights stored in the checkpoint.
-- Define a text and a few questions.
-- Iterate over the questions and build a sequence from the text and the current question, with the correct
-  model-specific separators token type ids and attention masks
-- Pass this sequence through the model. This outputs a range of scores across the entire sequence tokens (question and
-  text), for both the start and end positions.
-- Compute the softmax of the result to get probabilities over the tokens
-- Fetch the tokens from the identified start and stop values, convert those tokens to a string.
-- Print the results
+1. Instantiate a tokenizer and a model from the checkpoint name. The model is
+   identified as a BERT model and loads it with the weights stored in the
+   checkpoint.
+2. Define a text and a few questions.
+3. Iterate over the questions and build a sequence from the text and the current
+   question, with the correct model-specific separators token type ids and
+   attention masks.
+4. Pass this sequence through the model. This outputs a range of scores across
+   the entire sequence tokens (question and text), for both the start and end
+   positions.
+5. Compute the softmax of the result to get probabilities over the tokens.
+6. Fetch the tokens from the identified start and stop values, convert those tokens to a string.
+7. Print the results.
 
 .. code-block::
 
@@ -291,8 +299,8 @@ Here is an example of question answering using a model and a tokenizer. The proc
 Language Modeling
 ----------------------------------------------------
 
-Language modeling is the task of fitting a model to a corpus, which can be domain specific. All popular transformer
-based models are trained using a variant of language modeling, e.g. BERT with masked language modeling, GPT-2 with
+Language modeling is the task of fitting a model to a corpus, which can be domain specific. All popular transformer-based
+models are trained using a variant of language modeling, e.g. BERT with masked language modeling, GPT-2 with
 causal language modeling.
 
 Language modeling can be useful outside of pre-training as well, for example to shift the model distribution to be
@@ -305,7 +313,7 @@ Masked Language Modeling
 Masked language modeling is the task of masking tokens in a sequence with a masking token, and prompting the model to
 fill that mask with an appropriate token. This allows the model to attend to both the right context (tokens on the
 right of the mask) and the left context (tokens on the left of the mask). Such a training creates a strong basis
-for downstream tasks requiring bi-directional context such as SQuAD (question answering,
+for downstream tasks, requiring bi-directional context such as SQuAD (question answering,
 see `Lewis, Lui, Goyal et al. <https://arxiv.org/abs/1910.13461>`__, part 4.2).
 
 Here is an example of using pipelines to replace a mask from a sequence:
@@ -316,7 +324,7 @@ Here is an example of using pipelines to replace a mask from a sequence:
 
     >>> nlp = pipeline("fill-mask")
 
-This outputs the sequences with the mask filled, the confidence score as well as the token id in the tokenizer
+This outputs the sequences with the mask filled, the confidence score, and the token id in the tokenizer
 vocabulary:
 
 .. code-block::
@@ -349,17 +357,19 @@ vocabulary:
       'token': 17715,
       'token_str': 'Ġprototype'}]
 
-Here is an example doing masked language modeling using a model and a tokenizer. The process is the following:
+Here is an example of doing masked language modeling using a model and a tokenizer. The process is the following:
 
-- Instantiate a tokenizer and a model from the checkpoint name. The model is identified as a DistilBERT model and
-  loads it with the weights stored in the checkpoint.
-- Define a sequence with a masked token, placing the :obj:`tokenizer.mask_token` instead of a word.
-- Encode that sequence into IDs and find the position of the masked token in that list of IDs.
-- Retrieve the predictions at the index of the mask token: this tensor has the same size as the vocabulary, and the
-  values are the scores attributed to each token. The model gives higher score to tokens he deems probable in that
-  context.
-- Retrieve the top 5 tokens using the PyTorch :obj:`topk` or TensorFlow :obj:`top_k` methods.
-- Replace the mask token by the tokens and print the results
+1. Instantiate a tokenizer and a model from the checkpoint name. The model is
+   identified as a DistilBERT model and loads it with the weights stored in the
+   checkpoint.
+2. Define a sequence with a masked token, placing the :obj:`tokenizer.mask_token` instead of a word.
+3. Encode that sequence into a list of IDs and find the position of the masked token in that list.
+4. Retrieve the predictions at the index of the mask token: this tensor has the
+   same size as the vocabulary, and the values are the scores attributed to each
+   token. The model gives higher score to tokens it deems probable in that
+   context.
+5. Retrieve the top 5 tokens using the PyTorch :obj:`topk` or TensorFlow :obj:`top_k` methods.
+6. Replace the mask token by the tokens and print the results
 
 .. code-block::
 
@@ -375,7 +385,7 @@ Here is an example doing masked language modeling using a model and a tokenizer.
     >>> input = tokenizer.encode(sequence, return_tensors="pt")
     >>> mask_token_index = torch.where(input == tokenizer.mask_token_id)[1]
 
-    >>> token_logits = model(input)[0]
+    >>> token_logits = model(input).logits
     >>> mask_token_logits = token_logits[0, mask_token_index, :]
 
     >>> top_5_tokens = torch.topk(mask_token_logits, 5, dim=1).indices[0].tolist()
@@ -419,7 +429,7 @@ for generation tasks.
 
 Usually, the next token is predicted by sampling from the logits of the last hidden state the model produces from the input sequence.
 
-Here is an example using the tokenizer and model and leveraging the :func:`~transformers.PreTrainedModel.top_k_top_p_filtering` method to sample the next token following an input sequence of tokens.
+Here is an example of using the tokenizer and model and leveraging the :func:`~transformers.PreTrainedModel.top_k_top_p_filtering` method to sample the next token following an input sequence of tokens.
 
 .. code-block::
 
@@ -436,7 +446,7 @@ Here is an example using the tokenizer and model and leveraging the :func:`~tran
     >>> input_ids = tokenizer.encode(sequence, return_tensors="pt")
 
     >>> # get logits of last hidden state
-    >>> next_token_logits = model(input_ids)[0][:, -1, :]
+    >>> next_token_logits = model(input_ids).logits[:, -1, :]
 
     >>> # filter
     >>> filtered_next_token_logits = top_k_top_p_filtering(next_token_logits, top_k=50, top_p=1.0)
@@ -473,11 +483,11 @@ Here is an example using the tokenizer and model and leveraging the :func:`~tran
     >>> resulting_string = tokenizer.decode(generated.numpy().tolist()[0])
 
 
-This outputs a (hopefully) coherent next token following the original sequence, which is in our case is the word *has*:
+This outputs a (hopefully) coherent next token following the original sequence, which in our case is the word *has*:
 
 .. code-block::
 
-    print(resulting_string)
+    >>> print(resulting_string)
     Hugging Face is based in DUMBO, New York City, and has
 
 In the next section, we show how this functionality is leveraged in :func:`~transformers.PreTrainedModel.generate` to generate multiple tokens up to a user-defined length.
@@ -485,7 +495,7 @@ In the next section, we show how this functionality is leveraged in :func:`~tran
 Text Generation
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-In text generation (*a.k.a* *open-ended text generation*) the goal is to create a coherent portion of text that is a continuation from the given context. As an example, is it shown how *GPT-2* can be used in pipelines to generate text. As a default all models apply *Top-K* sampling when used in pipelines as configured in their respective configurations (see `gpt-2 config <https://s3.amazonaws.com/models.huggingface.co/bert/gpt2-config.json>`_ for example).
+In text generation (*a.k.a* *open-ended text generation*) the goal is to create a coherent portion of text that is a continuation from the given context. The following example shows how *GPT-2* can be used in pipelines to generate text. As a default all models apply *Top-K* sampling when used in pipelines, as configured in their respective configurations (see `gpt-2 config <https://s3.amazonaws.com/models.huggingface.co/bert/gpt2-config.json>`__ for example).
 
 .. code-block::
 
@@ -497,10 +507,10 @@ In text generation (*a.k.a* *open-ended text generation*) the goal is to create
 
 
 
-Here the model generates a random text with a total maximal length of *50* tokens from context *"As far as I am concerned, I will"*.
-The default arguments of ``PreTrainedModel.generate()`` can directly be overriden in the pipeline as is shown above for the argument ``max_length``.
+Here, the model generates a random text with a total maximal length of *50* tokens from context *"As far as I am concerned, I will"*.
+The default arguments of ``PreTrainedModel.generate()`` can be directly overriden in the pipeline, as is shown above for the argument ``max_length``.
 
-Here is an example for text generation using XLNet and its tokenzier. 
+Here is an example of text generation using ``XLNet`` and its tokenzier.
 
 .. code-block::
 
@@ -556,24 +566,27 @@ Here is an example for text generation using XLNet and its tokenzier.
 
 .. code-block::
 
-    print(generated)
+    >>> print(generated)
+    Today the weather is really nice and I am planning on anning on taking a nice...... of a great time!<eop>...............
 
-Text generation is currently possible with *GPT-2*, *OpenAi-GPT*, *CTRL*, *XLNet*, *Transfo-XL* and *Reformer* in PyTorch and for most models in Tensorflow as well. As can be seen in the example above *XLNet* and *Transfo-xl* often need to be padded to work well.
-GPT-2 is usually a good choice for *open-ended text generation* because it was trained on millions on webpages with a causal language modeling objective.
+Text generation is currently possible with *GPT-2*, *OpenAi-GPT*, *CTRL*, *XLNet*, *Transfo-XL* and *Reformer* in PyTorch and for most models in Tensorflow as well. As can be seen in the example above *XLNet* and *Transfo-XL* often need to be padded to work well.
+GPT-2 is usually a good choice for *open-ended text generation* because it was trained on millions of webpages with a causal language modeling objective.
 
-For more information on how to apply different decoding strategies for text generation, please also refer to our generation blog post `here <https://huggingface.co/blog/how-to-generate>`_.
+For more information on how to apply different decoding strategies for text generation, please also refer to our text generation blog post `here <https://huggingface.co/blog/how-to-generate>`__.
 
 
 Named Entity Recognition
 ----------------------------------------------------
 
-Named Entity Recognition (NER) is the task of classifying tokens according to a class, for example identifying a
+Named Entity Recognition (NER) is the task of classifying tokens according to a class, for example, identifying a
 token as a person, an organisation or a location.
 An example of a named entity recognition dataset is the CoNLL-2003 dataset, which is entirely based on that task.
-If you would like to fine-tune a model on an NER task, you may leverage the `ner/run_ner.py` (PyTorch),
-`ner/run_pl_ner.py` (leveraging pytorch-lightning) or the `ner/run_tf_ner.py` (TensorFlow) scripts.
+If you would like to fine-tune a model on an NER task, you may leverage the
+`run_ner.py <https://github.com/huggingface/transformers/tree/master/examples/token-classification/run_ner.py>`__ (PyTorch),
+`run_pl_ner.py <https://github.com/huggingface/transformers/tree/master/examples/token-classification/run_pl_ner.py>`__ (leveraging pytorch-lightning) or the
+`run_tf_ner.py <https://github.com/huggingface/transformers/tree/master/examples/token-classification/run_tf_ner.py>`__ (TensorFlow) scripts.
 
-Here is an example using the pipelines do to named entity recognition, trying to identify tokens as belonging to one
+Here is an example of using pipelines to do named entity recognition, specifically, trying to identify tokens as belonging to one
 of 9 classes:
 
 - O, Outside of a named entity
@@ -599,13 +612,12 @@ It leverages a fine-tuned model on CoNLL-2003, fine-tuned by `@stefan-it <https:
     ...            "close to the Manhattan Bridge which is visible from the window."
 
 
-This outputs a list of all words that have been identified as an entity from the 9 classes defined above. Here is the
+This outputs a list of all words that have been identified as one of the entities from the 9 classes defined above. Here are the
 expected results:
 
 .. code-block::
 
-    print(nlp(sequence))
-
+    >>> print(nlp(sequence))
     [
         {'word': 'Hu', 'score': 0.9995632767677307, 'entity': 'I-ORG'},
         {'word': '##gging', 'score': 0.9915938973426819, 'entity': 'I-ORG'},
@@ -621,22 +633,25 @@ expected results:
         {'word': 'Bridge', 'score': 0.990249514579773, 'entity': 'I-LOC'}
     ]
 
-Note how the words "Hugging Face" have been identified as an organisation, and "New York City", "DUMBO" and
+Note, how the tokens of the sequence "Hugging Face" have been identified as an organisation, and "New York City", "DUMBO" and
 "Manhattan Bridge" have been identified as locations.
 
-Here is an example doing named entity recognition using a model and a tokenizer. The process is the following:
+Here is an example of doing named entity recognition, using a model and a tokenizer. The process is the following:
 
-- Instantiate a tokenizer and a model from the checkpoint name. The model is identified as a BERT model and
-  loads it with the weights stored in the checkpoint.
-- Define the label list with which the model was trained on.
-- Define a sequence with known entities, such as "Hugging Face" as an organisation and "New York City" as a location.
-- Split words into tokens so that they can be mapped to the predictions. We use a small hack by firstly completely
-  encoding and decoding the sequence, so that we're left with a string that contains the special tokens.
-- Encode that sequence into IDs (special tokens are added automatically).
-- Retrieve the predictions by passing the input to the model and getting the first output. This results in a
-  distribution over the 9 possible classes for each token. We take the argmax to retrieve the most likely class
-  for each token.
-- Zip together each token with its prediction and print it.
+1. Instantiate a tokenizer and a model from the checkpoint name. The model is
+   identified as a BERT model and loads it with the weights stored in the
+   checkpoint.
+2. Define the label list with which the model was trained on.
+3. Define a sequence with known entities, such as "Hugging Face" as an organisation and "New York City" as a location.
+4. Split words into tokens so that they can be mapped to predictions. We use a
+   small hack by, first, completely encoding and decoding the sequence, so that
+   we're left with a string that contains the special tokens.
+5. Encode that sequence into IDs (special tokens are added automatically).
+6. Retrieve the predictions by passing the input to the model and getting the
+   first output. This results in a distribution over the 9 possible classes for
+   each token. We take the argmax to retrieve the most likely class for each
+   token.
+7. Zip together each token with its prediction and print it.
 
 .. code-block::
 
@@ -666,7 +681,7 @@ Here is an example doing named entity recognition using a model and a tokenizer.
     >>> tokens = tokenizer.tokenize(tokenizer.decode(tokenizer.encode(sequence)))
     >>> inputs = tokenizer.encode(sequence, return_tensors="pt")
 
-    >>> outputs = model(inputs)[0]
+    >>> outputs = model(inputs).logits
     >>> predictions = torch.argmax(outputs, dim=2)
     >>> ## TENSORFLOW CODE
     >>> from transformers import TFAutoModelForTokenClassification, AutoTokenizer
@@ -698,8 +713,8 @@ Here is an example doing named entity recognition using a model and a tokenizer.
     >>> predictions = tf.argmax(outputs, axis=2)
 
 
-This outputs a list of each token mapped to their prediction. Differently from the pipeline, here every token has
-a prediction as we didn't remove the "0" class which means that no particular entity was found on that token. The
+This outputs a list of each token mapped to its corresponding prediction. Differently from the pipeline, here every token has
+a prediction as we didn't remove the "0"th class, which means that no particular entity was found on that token. The
 following array should be the output:
 
 .. code-block::
@@ -710,13 +725,13 @@ following array should be the output:
 Summarization
 ----------------------------------------------------
 
-Summarization is the task of summarizing a text / an article into a shorter text.
+Summarization is the task of summarizing a document or an article into a shorter text.
 
 An example of a summarization dataset is the CNN / Daily Mail dataset, which consists of long news articles and was created for the task of summarization.
-If you would like to fine-tune a model on a summarization task, you may leverage the ``examples/summarization/bart/run_train.sh`` (leveraging pytorch-lightning) script.
+If you would like to fine-tune a model on a summarization task, various approaches are described in this
+`document <https://github.com/huggingface/transformers/blob/master/examples/seq2seq/README.md>`__.
 
-Here is an example using the pipelines do to summarization. 
-It leverages a Bart model that was fine-tuned on the CNN / Daily Mail data set.
+Here is an example of using the pipelines to do summarization. It leverages a Bart model that was fine-tuned on the CNN / Daily Mail data set.
 
 .. code-block::
 
@@ -743,8 +758,8 @@ It leverages a Bart model that was fine-tuned on the CNN / Daily Mail data set.
     ... If convicted, Barrientos faces up to four years in prison.  Her next court appearance is scheduled for May 18.
     ... """
 
-Because the summarization pipeline depends on the ``PretrainedModel.generate()`` method, we can override the default arguments 
-of ``PretrainedModel.generate()`` directly in the pipeline as is shown for ``max_length`` and ``min_length`` above.
+Because the summarization pipeline depends on the ``PretrainedModel.generate()`` method, we can override the default arguments
+of ``PretrainedModel.generate()`` directly in the pipeline for ``max_length`` and ``min_length`` as shown below.
 This outputs the following summary:
 
 .. code-block::
@@ -752,14 +767,15 @@ This outputs the following summary:
     >>> print(summarizer(ARTICLE, max_length=130, min_length=30, do_sample=False))
     [{'summary_text': 'Liana Barrientos, 39, is charged with two counts of "offering a false instrument for filing in the first degree" In total, she has been married 10 times, with nine of her marriages occurring between 1999 and 2002. She is believed to still be married to four men.'}]
 
-Here is an example doing summarization using a model and a tokenizer. The process is the following:
+Here is an example of doing summarization using a model and a tokenizer. The process is the following:
 
-- Instantiate a tokenizer and a model from the checkpoint name. Summarization is usually done using an encoder-decoder model, such as ``Bart`` or ``T5``.
-- Define the article that should be summarizaed.
-- Leverage the ``PretrainedModel.generate()`` method.
-- Add the T5 specific prefix "summarize: ".
+1. Instantiate a tokenizer and a model from the checkpoint name. Summarization is usually done using an encoder-decoder model, such as ``Bart`` or ``T5``.
+2. Define the article that should be summarized.
+3. Add the T5 specific prefix "summarize: ".
+4. Use the ``PretrainedModel.generate()`` method to generate the summary.
+
+In this example we use Google`s T5 model. Even though it was pre-trained only on a multi-task mixed dataset (including CNN / Daily Mail), it yields very good results.
 
-Here Google`s T5 model is used that was only pre-trained on a multi-task mixed data set (including CNN / Daily Mail), but nevertheless yields very good results.
 .. code-block::
 
     >>> ## PYTORCH CODE
@@ -786,12 +802,14 @@ Translation
 
 Translation is the task of translating a text from one language to another.
 
-An example of a translation dataset is the WMT English to German dataset, which has English sentences as the input data 
-and German sentences as the target data.
+An example of a translation dataset is the WMT English to German dataset, which has sentences in English as the input data
+and the corresponding sentences in German as the target data.
+If you would like to fine-tune a model on a translation task, various approaches are described in this
+`document <https://github.com/huggingface/transformers/blob/master/examples/seq2seq/README.md>`__.
 
-Here is an example using the pipelines do to translation. 
-It leverages a T5 model that was only pre-trained on a multi-task mixture dataset (including WMT), but yields impressive 
-translation results nevertheless.
+Here is an example of using the pipelines to do translation.
+It leverages a T5 model that was only pre-trained on a multi-task mixture dataset (including WMT), yet, yielding impressive
+translation results.
 
 .. code-block::
 
@@ -801,20 +819,15 @@ translation results nevertheless.
     >>> print(translator("Hugging Face is a technology company based in New York and Paris", max_length=40))
     [{'translation_text': 'Hugging Face ist ein Technologieunternehmen mit Sitz in New York und Paris.'}]
 
-Because the translation pipeline depends on the ``PretrainedModel.generate()`` method, we can override the default arguments 
+Because the translation pipeline depends on the ``PretrainedModel.generate()`` method, we can override the default arguments
 of ``PretrainedModel.generate()`` directly in the pipeline as is shown for ``max_length`` above.
-This outputs the following translation into German:
 
-::
+Here is an example of doing translation using a model and a tokenizer. The process is the following:
 
-  Hugging Face ist ein Technologieunternehmen mit Sitz in New York und Paris.
-  
-Here is an example doing translation using a model and a tokenizer. The process is the following:
-
-- Instantiate a tokenizer and a model from the checkpoint name. Summarization is usually done using an encoder-decoder model, such as ``Bart`` or ``T5``.
-- Define the article that should be summarizaed.
-- Leverage the ``PretrainedModel.generate()`` method.
-- Add the T5 specific prefix "translate English to German: "
+1. Instantiate a tokenizer and a model from the checkpoint name. Summarization is usually done using an encoder-decoder model, such as ``Bart`` or ``T5``.
+2. Define the article that should be summarizaed.
+3. Add the T5 specific prefix "translate English to German: "
+4. Use the ``PretrainedModel.generate()`` method to perform the translation.
 
 .. code-block::
 
@@ -826,10 +839,6 @@ Here is an example doing translation using a model and a tokenizer. The process
 
     >>> inputs = tokenizer.encode("translate English to German: Hugging Face is a technology company based in New York and Paris", return_tensors="pt")
     >>> outputs = model.generate(inputs, max_length=40, num_beams=4, early_stopping=True)
-
-    >>> print(outputs)
-    tensor([[    0, 11560,  3896,  8881,   229,   236,     3, 14366, 15377,   181,
-             11216,    16,   368,  1060,    64,  1919,     5]])
     >>> ## TENSORFLOW CODE
     >>> from transformers import TFAutoModelWithLMHead, AutoTokenizer
 
@@ -839,7 +848,9 @@ Here is an example doing translation using a model and a tokenizer. The process
     >>> inputs = tokenizer.encode("translate English to German: Hugging Face is a technology company based in New York and Paris", return_tensors="tf")
     >>> outputs = model.generate(inputs, max_length=40, num_beams=4, early_stopping=True)
 
-    >>> print(outputs)
-    tf.Tensor(
-    [[    0 11560  3896  8881   229   236     3 14366 15377   181 11216    16
-        368  1060    64  1919     5]], shape=(1, 17), dtype=int32)
+As with the pipeline example, we get the same translation:
+
+.. code-block::
+
+    >>> print(tokenizer.decode(outputs[0]))
+    Hugging Face ist ein Technologieunternehmen mit Sitz in New York und Paris.

docs/source/tokenizer_summary.rst

@@ -0,0 +1,243 @@
+Tokenizer summary
+-----------------
+
+In this page, we will have a closer look at tokenization. As we saw in
+:doc:`the preprocessing tutorial <preprocessing>`, tokenizing a text is splitting it into words or subwords, which then
+are converted to ids. The second part is pretty straightforward, here we will focus on the first part. More
+specifically, we will look at the three main different kinds of tokenizers used in 🤗 Transformers:
+:ref:`Byte-Pair Encoding (BPE) <byte-pair-encoding>`, :ref:`WordPiece <wordpiece>` and
+:ref:`SentencePiece <sentencepiece>`, and provide examples of models using each of those.
+
+Note that on each model page, you can look at the documentation of the associated tokenizer to know which of those
+algorithms the pretrained model used. For instance, if we look at :class:`~transformers.BertTokenizer`, we can see it's
+using :ref:`WordPiece <wordpiece>`.
+
+Introduction to tokenization
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Splitting a text in smaller chunks is a task that's harder than it looks, and there are multiple ways of doing it. For
+instance, let's look at the sentence "Don't you love 🤗 Transformers? We sure do." A first simple way of tokenizing
+this text is just to split it by spaces, which would give:
+
+::
+
+    ["Don't", "you", "love", "🤗", "Transformers?", "We", "sure", "do."]
+
+This is a nice first step, but if we look at the tokens "Transformers?" or "do.", we can see we can do better. Those
+will be different than the tokens "Transformers" and "do" for our model, so we should probably take the punctuation
+into account. This would give:
+
+::
+
+    ["Don", "'", "t", "you", "love", "🤗", "Transformers", "?", "We", "sure", "do", "."]
+
+which is better already. One thing that is annoying though is how it dealt with "Don't". "Don't" stands for do not, so
+it should probably be better tokenized as ``["Do", "n't"]``. This is where things start getting more complicated, and
+part of the reason each kind of model has its own tokenizer class. Depending on the rules we apply to split our texts
+into tokens, we'll get different tokenized versions of the same text. And of course, a given pretrained model won't
+perform properly if you don't use the exact same rules as the persons who pretrained it.
+
+`spaCy <https://spacy.io/>`__ and `Moses <http://www.statmt.org/moses/?n=Development.GetStarted>`__ are two popular
+rule-based tokenizers. On the text above, they'd output something like:
+
+::
+
+    ["Do", "n't", "you", "love", "🤗", "Transformers", "?", "We", "sure", "do", "."]
+
+Space/punctuation-tokenization and rule-based tokenization are both examples of word tokenization, which is splitting a
+sentence into words. While it's the most intuitive way to separate texts in smaller chunks, it can have a problem when
+you have a huge corpus: it usually yields a very big vocabulary (the set of all unique tokens used).
+:doc:`Transformer XL <model_doc/transformerxl>` for instance uses space/punctuation-tokenization, and has a vocabulary
+size of 267,735!
+
+A huge vocabulary size means a huge embedding matrix at the start of the model, which will cause memory problems.
+TransformerXL deals with it by using a special kind of embeddings called adaptive embeddings, but in general,
+transformers models rarely have a vocabulary size greater than 50,000, especially if they are trained on a single
+language.
+
+So if tokenizing on words is unsatisfactory, we could go on the opposite direction and simply tokenize on characters.
+While it's very simple and would save a lot of memory, this doesn't allow the model to learn representations of texts
+as meaningful as when using a word tokenization, leading to a loss of performance. So to get the best of both worlds,
+all transformers models use a hybrid between word-level and character-level tokenization called subword tokenization.
+
+Subword tokenization
+^^^^^^^^^^^^^^^^^^^^
+
+Subword tokenization algorithms rely on the principle that most common words should be left as is, but rare words
+should be decomposed in meaningful subword units. For instance "annoyingly" might be considered a rare word and
+decomposed as "annoying" and "ly". This is especially useful in agglutinative languages such as Turkish, where you can
+form (almost) arbitrarily long complex words by stringing together some subwords.
+
+This allows the model to keep a reasonable vocabulary while still learning useful representations for common words or
+subwords. This also enables the model to process words it has never seen before, by decomposing them into
+subwords it knows. For instance, the base :class:`~transformers.BertTokenizer` will tokenize "I have a new GPU!" like
+this:
+
+.. code-block::
+
+    >>> from transformers import BertTokenizer
+    >>> tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+    >>> tokenizer.tokenize("I have a new GPU!")
+    ['i', 'have', 'a', 'new', 'gp', '##u', '!']
+
+Since we are considering the uncased model, the sentence was lowercased first. Then all the words were present in the
+vocabulary of the tokenizer, except for "gpu", so the tokenizer split it in subwords it knows: "gp" and "##u". The "##"
+means that the rest of the token should be attached to the previous one, without space (for when we need to decode
+predictions and reverse the tokenization).
+
+Another example is when we use the base :class:`~transformers.XLNetTokenizer` to tokenize our previous text:
+
+.. code-block::
+
+    >>> from transformers import XLNetTokenizer
+    >>> tokenizer = XLNetTokenizer.from_pretrained('xlnet-base-cased')
+    >>> tokenizer.tokenize("Don't you love 🤗 Transformers? We sure do.")
+    ['▁Don', "'", 't', '▁you', '▁love', '▁', '🤗', '▁', 'Transform', 'ers', '?', '▁We', '▁sure', '▁do', '.']
+
+We'll get back to the meaning of those '▁' when we look at :ref:`SentencePiece <sentencepiece>` but you can see
+Transformers has been split into "Transform" and "ers".
+
+Let's now look at how the different subword tokenization algorithms work. Note that they all rely on some form of
+training which is usually done on the corpus the corresponding model will be trained on.
+
+.. _byte-pair-encoding:
+
+Byte-Pair Encoding
+~~~~~~~~~~~~~~~~~~
+
+Byte-Pair Encoding was introduced in `this paper <https://arxiv.org/abs/1508.07909>`__. It relies on a pretokenizer
+splitting the training data into words, which can be a simple space tokenization
+(:doc:`GPT-2 <model_doc/gpt2>` and :doc:`Roberta <model_doc/roberta>` uses this for instance) or a rule-based tokenizer
+(:doc:`XLM <model_doc/xlm>` use Moses for most languages, as does :doc:`FlauBERT <model_doc/flaubert>`),
+
+:doc:`GPT <model_doc/gpt>` uses Spacy and ftfy, and counts the frequency of each word in the training corpus.
+
+It then begins from the list of all characters, and will learn merge rules to form a new token from two symbols in the
+vocabulary until it has learned a vocabulary of the desired size (this is a hyperparameter to pick).
+
+Let's say that after the pre-tokenization we have the following words (the number indicating the frequency of each
+word):
+
+::
+
+    ('hug', 10), ('pug', 5), ('pun', 12), ('bun', 4), ('hugs', 5)
+
+Then the base vocabulary is ['b', 'g', 'h', 'n', 'p', 's', 'u'] and all our words are first split by character:
+
+::
+
+    ('h' 'u' 'g', 10), ('p' 'u' 'g', 5), ('p' 'u' 'n', 12), ('b' 'u' 'n', 4), ('h' 'u' 'g' 's', 5)
+
+We then take each pair of symbols and look at the most frequent. For instance 'hu' is present `10 + 5 = 15` times (10
+times in the 10 occurrences of 'hug', 5 times in the 5 occurrences of 'hugs'). The most frequent here is 'ug', present
+`10 + 5 + 5 = 20` times in total. So the first merge rule the tokenizer learns is to group all 'u' and 'g' together
+then it adds 'ug' to the vocabulary. Our corpus then becomes
+
+::
+
+    ('h' 'ug', 10), ('p' 'ug', 5), ('p' 'u' 'n', 12), ('b' 'u' 'n', 4), ('h' 'ug' 's', 5)
+
+and we continue by looking at the next most common pair of symbols. It's 'un', present 16 times, so we merge those two
+and add 'un' to the vocabulary. Then it's 'hug' (as 'h' + 'ug'), present 15 times, so we merge those two and add 'hug'
+to the vocabulary.
+
+At this stage, the vocabulary is ``['b', 'g', 'h', 'n', 'p', 's', 'u', 'ug', 'un', 'hug']`` and our corpus is
+represented as
+
+::
+
+    ('hug', 10), ('p' 'ug', 5), ('p' 'un', 12), ('b' 'un', 4), ('hug' 's', 5)
+
+If we stop there, the tokenizer can apply the rules it learned to new words (as long as they don't contain characters that
+were not in the base vocabulary). For instance 'bug' would be tokenized as ``['b', 'ug']`` but mug would be tokenized as
+``['<unk>', 'ug']`` since the 'm' is not in the base vocabulary. This doesn't happen to letters in general (since the
+base corpus uses all of them), but to special characters like emojis.
+
+As we said before, the vocabulary size (which is the base vocabulary size + the number of merges) is a hyperparameter
+to choose. For instance :doc:`GPT <model_doc/gpt>` has a vocabulary size of 40,478 since they have 478 base characters
+and chose to stop the training of the tokenizer at 40,000 merges.
+
+Byte-level BPE
+^^^^^^^^^^^^^^
+
+To deal with the fact the base vocabulary needs to get all base characters, which can be quite big if one allows for
+all unicode characters, the
+`GPT-2 paper <https://cdn.openai.com/better-language-models/language_models_are_unsupervised_multitask_learners.pdf>`__
+introduces a clever trick, which is to use bytes as the base vocabulary (which gives a size of 256). With some
+additional rules to deal with punctuation, this manages to be able to tokenize every text without needing an unknown
+token. For instance, the :doc:`GPT-2 model <model_doc/gpt>` has a vocabulary size of 50,257, which corresponds to the
+256 bytes base tokens, a special end-of-text token and the symbols learned with 50,000 merges.
+
+.. _wordpiece:
+
+WordPiece
+=========
+
+WordPiece is the subword tokenization algorithm used for :doc:`BERT <model_doc/bert>` (as well as
+:doc:`DistilBERT <model_doc/distilbert>` and :doc:`Electra <model_doc/electra>`) and was outlined in
+`this paper <https://static.googleusercontent.com/media/research.google.com/ja//pubs/archive/37842.pdf>`__. It relies
+on the same base as BPE, which is to initialize the vocabulary to every character present in the corpus and
+progressively learn a given number of merge rules, the difference is that it doesn't choose the pair that is the most
+frequent but the one that will maximize the likelihood on the corpus once merged.
+
+What does this mean? Well, in the previous example, it means we would only merge 'u' and 'g' if the probability of
+having 'ug' divided by the probability of having 'u' then 'g' is greater than for any other pair of symbols. It's
+subtly different from what BPE does in the sense that it evaluates what it "loses" by merging two symbols and makes
+sure it's `worth it`.
+
+.. _unigram:
+
+Unigram
+=======
+
+Unigram is a subword tokenization algorithm introduced in `this paper <https://arxiv.org/pdf/1804.10959.pdf>`__.
+Instead of starting with a group of base symbols and learning merges with some rule, like BPE or WordPiece, it starts
+from a large vocabulary (for instance, all pretokenized words and the most common substrings) that it will trim down
+progressively. It's not used directly for any of the pretrained models in the library, but it's used in conjunction
+with :ref:`SentencePiece <sentencepiece>`.
+
+More specifically, at a given step, unigram computes a loss from the corpus we have and the current vocabulary, then,
+for each subword, evaluate how much the loss would augment if the subword was removed from the vocabulary. It then
+sorts the subwords by this quantity (that represents how worse the loss becomes if the token is removed) and removes
+all the worst p tokens (for instance p could be 10% or 20%). It then repeats the process until the vocabulary has
+reached the desired size, always keeping the base characters (to be able to tokenize any word written with them, like
+BPE or WordPiece).
+
+Contrary to BPE and WordPiece that work out rules in a certain order that you can then apply in the same order when
+tokenizing new text, Unigram will have several ways of tokenizing a new text. For instance, if it ends up with the
+vocabulary
+
+::
+
+    ['b', 'g', 'h', 'n', 'p', 's', 'u', 'ug', 'un', 'hug']
+
+we had before, it could tokenize "hugs" as ``['hug', 's']``, ``['h', 'ug', 's']`` or ``['h', 'u', 'g', 's']``. So which
+one choose? On top of saving the vocabulary, the trained tokenizer will save the probability of each token in the
+training corpus. You can then give a probability to each tokenization (which is the product of the probabilities of the
+tokens forming it) and pick the most likely one (or if you want to apply some data augmentation, you could sample one
+of the tokenization according to their probabilities).
+
+Those probabilities define the loss that trains the tokenizer: if our corpus consists of the
+words :math:`x_{1}, \dots, x_{N}` and if for the word :math:`x_{i}` we note :math:`S(x_{i})` the set of all possible
+tokenizations of :math:`x_{i}` (with the current vocabulary), then the loss is defined as
+
+.. math::
+    \mathcal{L} = -\sum_{i=1}^{N} \log \left ( \sum_{x \in S(x_{i})} p(x) \right )
+
+.. _sentencepiece:
+
+SentencePiece
+=============
+
+All the methods we have been looking at so far required some form of pretokenization, which has a central problem: not
+all languages use spaces to separate words. This is a problem :doc:`XLM <model_doc/xlm>` solves by using specific
+pretokenizers for each of those languages (in this case, Chinese, Japanese and Thai). To solve this problem,
+SentencePiece (introduced in `this paper <https://arxiv.org/pdf/1808.06226.pdf>`__) treats the input as a raw stream,
+includes the space in the set of characters to use, then uses BPE or unigram to construct the appropriate vocabulary.
+
+That's why in the example we saw before using :class:`~transformers.XLNetTokenizer` (which uses SentencePiece), we had
+the '▁' character, that represents space. Decoding a tokenized text is then super easy: we just have to concatenate
+all of them together and replace '▁' with space.
+
+All transformers models in the library that use SentencePiece use it with unigram. Examples of models using it are
+:doc:`ALBERT <model_doc/albert>`, :doc:`XLNet <model_doc/xlnet>` or the :doc:`Marian framework <model_doc/marian>`.

docs/source/torchscript.rst

@@ -1,135 +0,0 @@
-TorchScript
-================================================
-
-.. note::
-    This is the very beginning of our experiments with TorchScript and we are still exploring its capabilities
-    with variable-input-size models. It is a focus of interest to us and we will deepen our analysis in upcoming
-    releases, with more code examples, a more flexible implementation, and benchmarks comparing python-based codes
-    with compiled TorchScript.
-
-
-According to Pytorch's documentation: "TorchScript is a way to create serializable and optimizable models from PyTorch code".
-Pytorch's two modules `JIT and TRACE <https://pytorch.org/docs/stable/jit.html>`_ allow the developer to export
-their model to be re-used in other programs, such as efficiency-oriented C++ programs.
-
-We have provided an interface that allows the export of 🤗 Transformers models to TorchScript so that they can
-be reused in a different environment than a Pytorch-based python program. Here we explain how to use our models so that
-they can be exported, and what to be mindful of when using these models with TorchScript.
-
-Exporting a model needs two things:
-
-* dummy inputs to execute a model forward pass.
-* the model needs to be instantiated with the ``torchscript`` flag.
-
-These necessities imply several things developers should be careful about. These are detailed below.
-
-
-Implications
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-TorchScript flag and tied weights
-------------------------------------------------
-This flag is necessary because most of the language models in this repository have tied weights between their
-``Embedding`` layer and their ``Decoding`` layer. TorchScript does not allow the export of models that have tied weights,
-it is therefore necessary to untie the weights beforehand.
-
-This implies that models instantiated with the ``torchscript`` flag have their ``Embedding`` layer and ``Decoding`` layer
-separate, which means that they should not be trained down the line. Training would de-synchronize the two layers,
-leading to unexpected results.
-
-This is not the case for models that do not have a Language Model head, as those do not have tied weights. These models
-can be safely exported without the ``torchscript`` flag.
-
-Dummy inputs and standard lengths
-------------------------------------------------
-
-The dummy inputs are used to do a model forward pass. While the inputs' values are propagating through the layers,
-Pytorch keeps track of the different operations executed on each tensor. These recorded operations are then used
-to create the "trace" of the model.
-
-The trace is created relatively to the inputs' dimensions. It is therefore constrained by the dimensions of the dummy
-input, and will not work for any other sequence length or batch size. When trying with a different size, an error such
-as:
-
-``The expanded size of the tensor (3) must match the existing size (7) at non-singleton dimension 2``
-
-will be raised. It is therefore recommended to trace the model with a dummy input size at least as large as the largest
-input that will be fed to the model during inference. Padding can be performed to fill the missing values. As the model
-will have been traced with a large input size however, the dimensions of the different matrix will be large as well,
-resulting in more calculations.
-
-It is recommended to be careful of the total number of operations done on each input and to follow performance closely
-when exporting varying sequence-length models.
-
-Using TorchScript in Python
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-Below are examples of using the Python to save, load models as well as how to use the trace for inference.
-
-Saving a model
-------------------------------------------------
-
-This snippet shows how to use TorchScript to export a ``BertModel``. Here the ``BertModel`` is instantiated
-according to a ``BertConfig`` class and then saved to disk under the filename ``traced_bert.pt``
-
-.. code-block:: python
-
-    from transformers import BertModel, BertTokenizer, BertConfig
-    import torch
-
-    enc = BertTokenizer.from_pretrained("bert-base-uncased")
-
-    # Tokenizing input text
-    text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
-    tokenized_text = enc.tokenize(text)
-
-    # Masking one of the input tokens
-    masked_index = 8
-    tokenized_text[masked_index] = '[MASK]'
-    indexed_tokens = enc.convert_tokens_to_ids(tokenized_text)
-    segments_ids = [0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1]
-
-    # Creating a dummy input
-    tokens_tensor = torch.tensor([indexed_tokens])
-    segments_tensors = torch.tensor([segments_ids])
-    dummy_input = [tokens_tensor, segments_tensors]
-
-    # Initializing the model with the torchscript flag
-    # Flag set to True even though it is not necessary as this model does not have an LM Head.
-    config = BertConfig(vocab_size_or_config_json_file=32000, hidden_size=768,
-        num_hidden_layers=12, num_attention_heads=12, intermediate_size=3072, torchscript=True)
-
-    # Instantiating the model
-    model = BertModel(config)
-
-    # The model needs to be in evaluation mode
-    model.eval()
-
-    # If you are instantiating the model with `from_pretrained` you can also easily set the TorchScript flag
-    model = BertModel.from_pretrained("bert-base-uncased", torchscript=True)
-
-    # Creating the trace
-    traced_model = torch.jit.trace(model, [tokens_tensor, segments_tensors])
-    torch.jit.save(traced_model, "traced_bert.pt")
-
-Loading a model
-------------------------------------------------
-
-This snippet shows how to load the ``BertModel`` that was previously saved to disk under the name ``traced_bert.pt``.
-We are re-using the previously initialised ``dummy_input``.
-
-.. code-block:: python
-
-    loaded_model = torch.jit.load("traced_model.pt")
-    loaded_model.eval()
-
-    all_encoder_layers, pooled_output = loaded_model(dummy_input)
-
-Using a traced model for inference
-------------------------------------------------
-
-Using the traced model for inference is as simple as using its ``__call__`` dunder method:
-
-.. code-block:: python
-
-    traced_model(tokens_tensor, segments_tensors)

docs/source/training.rst

@@ -16,10 +16,10 @@ TF2, and focus specifically on the nuances and tools for training models in
 
 Sections:
 
-  * :ref:`pytorch`
-  * :ref:`tensorflow`
-  * :ref:`trainer`
-  * :ref:`additional-resources`
+  - :ref:`pytorch`
+  - :ref:`tensorflow`
+  - :ref:`trainer`
+  - :ref:`additional-resources`
 
 .. _pytorch:
 
@@ -39,7 +39,7 @@ of the specified model are used to initialize the model. The
 library also includes a number of task-specific final layers or 'heads' whose
 weights are instantiated randomly when not present in the specified
 pre-trained model. For example, instantiating a model with
-``BertForSequenceClassification.from_pretrained('bert-base-uncased', num_classes=2)``
+``BertForSequenceClassification.from_pretrained('bert-base-uncased', num_labels=2)``
 will create a BERT model instance with encoder weights copied from the
 ``bert-base-uncased`` model and a randomly initialized sequence
 classification head on top of the encoder with an output size of 2. Models
@@ -49,7 +49,7 @@ put it in train mode.
 .. code-block:: python
 
     from transformers import BertForSequenceClassification
-    model = BertForSequenceClassification.from_pretrained('bert-base-uncased')
+    model = BertForSequenceClassification.from_pretrained('bert-base-uncased', return_dict=True)
     model.train()
 
 This is useful because it allows us to make use of the pre-trained BERT
@@ -99,7 +99,7 @@ backwards pass and update the weights:
 
     labels = torch.tensor([1,0]).unsqueeze(0)
     outputs = model(input_ids, attention_mask=attention_mask, labels=labels)
-    loss = outputs[0]
+    loss = outputs.loss
     loss.backward()
     optimizer.step()
 
@@ -111,7 +111,7 @@ The following is equivalent to the previous example:
     from torch.nn import functional as F
     labels = torch.tensor([1,0]).unsqueeze(0)
     outputs = model(input_ids, attention_mask=attention_mask)
-    loss = F.cross_entropy(labels, outputs[0])
+    loss = F.cross_entropy(labels, outputs.logitd)
     loss.backward()
     optimizer.step()
 
@@ -131,7 +131,6 @@ Then all we have to do is call ``scheduler.step()`` after ``optimizer.step()``.
 
 .. code-block:: python
 
-    ...
     loss.backward()
     optimizer.step()
     scheduler.step()
@@ -151,7 +150,7 @@ the encoder parameters, which can be accessed with the ``base_model``
 submodule on any task-specific model in the library:
 
 .. code-block:: python
-   
+
     for param in model.base_model.parameters():
         param.requires_grad = False
 
@@ -182,6 +181,7 @@ the pretrained tokenizer name.
 .. code-block:: python
 
     from transformers import BertTokenizer, glue_convert_examples_to_features
+    import tensorflow as tf
     import tensorflow_datasets as tfds
     tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
     data = tfds.load('glue/mrpc')
@@ -191,7 +191,7 @@ the pretrained tokenizer name.
 The model can then be compiled and trained as any Keras model:
 
 .. code-block:: python
-    
+
     optimizer = tf.keras.optimizers.Adam(learning_rate=3e-5)
     loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
     model.compile(optimizer=optimizer, loss=loss)
@@ -272,7 +272,7 @@ optimize.
 :func:`~transformers.Trainer` uses a built-in default function to collate
 batches and prepare them to be fed into the model. If needed, you can also
 use the ``data_collator`` argument to pass your own collator function which
-takes in the data in the format provides by your dataset and returns a
+takes in the data in the format provided by your dataset and returns a
 batch ready to be fed into the model. Note that
 :func:`~transformers.TFTrainer` expects the passed datasets to be dataset
 objects from ``tensorflow_datasets``.
@@ -282,7 +282,7 @@ your own ``compute_metrics`` function and pass it to the trainer.
 
 .. code-block:: python
 
-    from sklearn.metrics import precision_recall_fscore_support
+    from sklearn.metrics import accuracy_score, precision_recall_fscore_support
 
     def compute_metrics(pred):
         labels = pred.label_ids
@@ -305,19 +305,14 @@ launching tensorboard in your specified ``logging_dir`` directory.
 Additional resources
 ^^^^^^^^^^^^^^^^^^^^
 
-    * `A lightweight colab demo
-      <https://colab.research.google.com/drive/1-JIJlao4dI-Ilww_NnTc0rxtp-ymgDgM?usp=sharing>`_
-      which uses ``Trainer`` for IMDb sentiment classification.
+- `A lightweight colab demo <https://colab.research.google.com/drive/1-JIJlao4dI-Ilww_NnTc0rxtp-ymgDgM?usp=sharing>`_
+  which uses ``Trainer`` for IMDb sentiment classification.
 
-    * `🤗 Transformers Examples <https://github.com/huggingface/transformers/tree/master/examples>`_
-      including scripts for training and fine-tuning on GLUE, SQuAD, and
-      several other tasks.
+- `🤗 Transformers Examples <https://github.com/huggingface/transformers/tree/master/examples>`_
+  including scripts for training and fine-tuning on GLUE, SQuAD, and several other tasks.
 
-    * `How to train a language model
-      <https://colab.research.google.com/github/huggingface/blog/blob/master/notebooks/01_how_to_train.ipynb>`_,
-      a detailed colab notebook which uses ``Trainer`` to train a masked
-      language model from scratch on Esperanto.
+- `How to train a language model <https://colab.research.google.com/github/huggingface/blog/blob/master/notebooks/01_how_to_train.ipynb>`_,
+  a detailed colab notebook which uses ``Trainer`` to train a masked language model from scratch on Esperanto.
 
-    * `🤗 Transformers Notebooks <./notebooks.html>`_ which contain dozens
-      of example notebooks from the community for training and using
-      🤗 Transformers on a variety of tasks.
+- `🤗 Transformers Notebooks <notebooks.html>`_ which contain dozens of example notebooks from the community for
+  training and using 🤗 Transformers on a variety of tasks.

examples/README.md

@@ -1,7 +1,7 @@
-## Examples
+# Examples
 
 Version 2.9 of 🤗 Transformers introduces a new [`Trainer`](https://github.com/huggingface/transformers/blob/master/src/transformers/trainer.py) class for PyTorch, and its equivalent [`TFTrainer`](https://github.com/huggingface/transformers/blob/master/src/transformers/trainer_tf.py) for TF 2.
-Running the examples requires PyTorch 1.3.1+ or TensorFlow 2.1+.
+Running the examples requires PyTorch 1.3.1+ or TensorFlow 2.2+.
 
 Here is the list of all our examples:
 - **grouped by task** (all official examples work for multiple models)
@@ -13,7 +13,7 @@ Here is the list of all our examples:
 This is still a work-in-progress – in particular documentation is still sparse – so please **contribute improvements/pull requests.**
 
 
-# The Big Table of Tasks
+## The Big Table of Tasks
 
 | Task | Example datasets | Trainer support | TFTrainer support | pytorch-lightning | Colab
 |---|---|:---:|:---:|:---:|:---:|
@@ -21,7 +21,7 @@ This is still a work-in-progress – in particular documentation is still sparse
 | [**`text-classification`**](https://github.com/huggingface/transformers/tree/master/examples/text-classification)   | GLUE, XNLI      | ✅ | ✅ | ✅ | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/master/notebooks/trainer/01_text_classification.ipynb)
 | [**`token-classification`**](https://github.com/huggingface/transformers/tree/master/examples/token-classification) | CoNLL NER       | ✅ | ✅ | ✅ | -
 | [**`multiple-choice`**](https://github.com/huggingface/transformers/tree/master/examples/multiple-choice)           | SWAG, RACE, ARC | ✅ | ✅ | -  | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ViktorAlm/notebooks/blob/master/MPC_GPU_Demo_for_TF_and_PT.ipynb)
-| [**`question-answering`**](https://github.com/huggingface/transformers/tree/master/examples/question-answering)     | SQuAD           | -  | ✅ | -  | -
+| [**`question-answering`**](https://github.com/huggingface/transformers/tree/master/examples/question-answering)     | SQuAD           | ✅ | ✅ | -  | -
 | [**`text-generation`**](https://github.com/huggingface/transformers/tree/master/examples/text-generation)           | -               | n/a | n/a | n/a | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/master/notebooks/02_how_to_generate.ipynb)
 | [**`distillation`**](https://github.com/huggingface/transformers/tree/master/examples/distillation)       | All               | -  | -  | -  | -
 | [**`summarization`**](https://github.com/huggingface/transformers/tree/master/examples/seq2seq)     | CNN/Daily Mail    | -  | -  | ✅  | -
@@ -78,3 +78,50 @@ python examples/xla_spawn.py --num_cores 8 \
 ```
 
 Feedback and more use cases and benchmarks involving TPUs are welcome, please share with the community.
+
+## Logging & Experiment tracking
+
+You can easily log and monitor your runs code. The following are currently supported:
+
+* [TensorBoard](https://www.tensorflow.org/tensorboard)
+* [Weights & Biases](https://docs.wandb.com/library/integrations/huggingface)
+* [Comet ML](https://www.comet.ml/docs/python-sdk/huggingface/)
+
+### Weights & Biases
+
+To use Weights & Biases, install the wandb package with:
+
+```bash
+pip install wandb
+```
+
+Then log in the command line:
+
+```bash
+wandb login
+```
+
+If you are in Jupyter or Colab, you should login with:
+
+```python
+import wandb
+wandb.login()
+```
+
+Whenever you use `Trainer` or `TFTrainer` classes, your losses, evaluation metrics, model topology and gradients (for `Trainer` only) will automatically be logged.
+
+When using 🤗 Transformers with PyTorch Lightning, runs can be tracked through `WandbLogger`. Refer to related [documentation & examples](https://docs.wandb.com/library/integrations/lightning).
+
+### Comet.ml
+
+To use `comet_ml`, install the Python package with:
+
+```bash
+pip install comet_ml
+```
+
+or if in a Conda environment:
+
+```bash
+conda install -c comet_ml -c anaconda -c conda-forge comet_ml
+```

examples/adversarial/utils_hans.py

@@ -20,8 +20,8 @@ from dataclasses import dataclass
 from typing import List, Optional, Union
 
 import tqdm
-from filelock import FileLock
 
+from filelock import FileLock
 from transformers import (
     BartTokenizer,
     BartTokenizerFast,
@@ -112,7 +112,10 @@ if is_torch_available():
             cached_features_file = os.path.join(
                 data_dir,
                 "cached_{}_{}_{}_{}".format(
-                    "dev" if evaluate else "train", tokenizer.__class__.__name__, str(max_seq_length), task,
+                    "dev" if evaluate else "train",
+                    tokenizer.__class__.__name__,
+                    str(max_seq_length),
+                    task,
                 ),
             )
             label_list = processor.get_labels()
@@ -255,7 +258,11 @@ class HansProcessor(DataProcessor):
         return self._create_examples(self._read_tsv(os.path.join(data_dir, "heuristics_evaluation_set.txt")), "dev")
 
     def get_labels(self):
-        """See base class."""
+        """See base class.
+        Note that we follow the standard three labels for MNLI
+        (see :class:`~transformers.data.processors.utils.MnliProcessor`)
+        but the HANS evaluation groups `contradiction` and `neutral` into `non-entailment` (label 0) while
+        `entailment` is label 1."""
         return ["contradiction", "entailment", "neutral"]
 
     def _create_examples(self, lines, set_type):
@@ -268,13 +275,16 @@ class HansProcessor(DataProcessor):
             text_a = line[5]
             text_b = line[6]
             pairID = line[7][2:] if line[7].startswith("ex") else line[7]
-            label = line[-1]
+            label = line[0]
             examples.append(InputExample(guid=guid, text_a=text_a, text_b=text_b, label=label, pairID=pairID))
         return examples
 
 
 def hans_convert_examples_to_features(
-    examples: List[InputExample], label_list: List[str], max_length: int, tokenizer: PreTrainedTokenizer,
+    examples: List[InputExample],
+    label_list: List[str],
+    max_length: int,
+    tokenizer: PreTrainedTokenizer,
 ):
     """
     Loads a data file into a list of ``InputFeatures``

examples/benchmarking/README.md

@@ -0,0 +1,10 @@
+# 🤗 Benchmark results
+
+Here, you can find a list of the different benchmark results created by the community.
+
+If you would like to list benchmark results on your favorite models of the [model hub](https://huggingface.co/models) here, please open a Pull Request and add it below.
+
+| Benchmark description | Results | Environment info |      Author      |
+|:----------|:-------------|:-------------|------:|
+| PyTorch Benchmark on inference for `bert-base-cased` |[memory](https://github.com/patrickvonplaten/files_to_link_to/blob/master/bert_benchmark/inference_memory.csv) | [env](https://github.com/patrickvonplaten/files_to_link_to/blob/master/bert_benchmark/env.csv) | [Partick von Platen](https://github.com/patrickvonplaten) | 
+| PyTorch Benchmark on inference for `bert-base-cased` |[time](https://github.com/patrickvonplaten/files_to_link_to/blob/master/bert_benchmark/inference_time.csv) | [env](https://github.com/patrickvonplaten/files_to_link_to/blob/master/bert_benchmark/env.csv) | [Partick von Platen](https://github.com/patrickvonplaten) | 

examples/benchmarking/plot_csv_file.py

@@ -20,7 +20,9 @@ class PlotArguments:
     Arguments pertaining to which model/config/tokenizer we are going to fine-tune, or train from scratch.
     """
 
-    csv_file: str = field(metadata={"help": "The csv file to plot."},)
+    csv_file: str = field(
+        metadata={"help": "The csv file to plot."},
+    )
     plot_along_batch: bool = field(
         default=False,
         metadata={"help": "Whether to plot along batch size or sequence lengh. Defaults to sequence length."},
@@ -30,7 +32,8 @@ class PlotArguments:
         metadata={"help": "Whether the csv file has time results or memory results. Defaults to memory results."},
     )
     no_log_scale: bool = field(
-        default=False, metadata={"help": "Disable logarithmic scale when plotting"},
+        default=False,
+        metadata={"help": "Disable logarithmic scale when plotting"},
     )
     is_train: bool = field(
         default=False,
@@ -39,7 +42,8 @@ class PlotArguments:
         },
     )
     figure_png_file: Optional[str] = field(
-        default=None, metadata={"help": "Filename under which the plot will be saved. If unused no plot is saved."},
+        default=None,
+        metadata={"help": "Filename under which the plot will be saved. If unused no plot is saved."},
     )
     short_model_names: Optional[List[str]] = list_field(
         default=None, metadata={"help": "List of model names that are used instead of the ones in the csv file."}

examples/bert-loses-patience/pabee/modeling_pabee_albert.py

@@ -101,30 +101,30 @@ class AlbertModelWithPabee(AlbertModel):
         regression=False,
     ):
         r"""
-    Return:
-        :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
-        last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
-            Sequence of hidden-states at the output of the last layer of the model.
-        pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(batch_size, hidden_size)`):
-            Last layer hidden-state of the first token of the sequence (classification token)
-            further processed by a Linear layer and a Tanh activation function. The Linear
-            layer weights are trained from the next sentence prediction (classification)
-            objective during pre-training.
+        Return:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
+            last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
+                Sequence of hidden-states at the output of the last layer of the model.
+            pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(batch_size, hidden_size)`):
+                Last layer hidden-state of the first token of the sequence (classification token)
+                further processed by a Linear layer and a Tanh activation function. The Linear
+                layer weights are trained from the next sentence prediction (classification)
+                objective during pre-training.
 
-            This output is usually *not* a good summary
-            of the semantic content of the input, you're often better with averaging or pooling
-            the sequence of hidden-states for the whole input sequence.
-        hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
-            of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+                This output is usually *not* a good summary
+                of the semantic content of the input, you're often better with averaging or pooling
+                the sequence of hidden-states for the whole input sequence.
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
 
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
-            :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
 
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
-            heads.
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
         """
 
         if input_ids is not None and inputs_embeds is not None:
@@ -157,7 +157,10 @@ class AlbertModelWithPabee(AlbertModel):
             res = []
             for i in range(self.config.num_hidden_layers):
                 encoder_outputs = self.encoder.adaptive_forward(
-                    encoder_outputs, current_layer=i, attention_mask=extended_attention_mask, head_mask=head_mask,
+                    encoder_outputs,
+                    current_layer=i,
+                    attention_mask=extended_attention_mask,
+                    head_mask=head_mask,
                 )
 
                 pooled_output = self.pooler_activation(self.pooler(encoder_outputs[0][:, 0]))
@@ -174,7 +177,10 @@ class AlbertModelWithPabee(AlbertModel):
             for i in range(self.config.num_hidden_layers):
                 calculated_layer_num += 1
                 encoder_outputs = self.encoder.adaptive_forward(
-                    encoder_outputs, current_layer=i, attention_mask=extended_attention_mask, head_mask=head_mask,
+                    encoder_outputs,
+                    current_layer=i,
+                    attention_mask=extended_attention_mask,
+                    head_mask=head_mask,
                 )
 
                 pooled_output = self.pooler_activation(self.pooler(encoder_outputs[0][:, 0]))
@@ -236,42 +242,42 @@ class AlbertForSequenceClassificationWithPabee(AlbertPreTrainedModel):
         labels=None,
     ):
         r"""
-        labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
-            Labels for computing the sequence classification/regression loss.
-            Indices should be in ``[0, ..., config.num_labels - 1]``.
-            If ``config.num_labels == 1`` a regression loss is computed (Mean-Square loss),
-            If ``config.num_labels > 1`` a classification loss is computed (Cross-Entropy).
+            labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
+                Labels for computing the sequence classification/regression loss.
+                Indices should be in ``[0, ..., config.num_labels - 1]``.
+                If ``config.num_labels == 1`` a regression loss is computed (Mean-Square loss),
+                If ``config.num_labels > 1`` a classification loss is computed (Cross-Entropy).
 
-    Returns:
-        :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
-        loss: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
-            Classification (or regression if config.num_labels==1) loss.
-        logits ``torch.FloatTensor`` of shape ``(batch_size, config.num_labels)``
-            Classification (or regression if config.num_labels==1) scores (before SoftMax).
-        hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
-            of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+        Returns:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
+            loss: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+                Classification (or regression if config.num_labels==1) loss.
+            logits ``torch.FloatTensor`` of shape ``(batch_size, config.num_labels)``
+                Classification (or regression if config.num_labels==1) scores (before SoftMax).
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
 
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
-            :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
 
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
-            heads.
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
 
-        Examples::
+            Examples::
 
-            from transformers import AlbertTokenizer
-            from pabee import AlbertForSequenceClassificationWithPabee
-            import torch
+                from transformers import AlbertTokenizer
+                from pabee import AlbertForSequenceClassificationWithPabee
+                import torch
 
-            tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
-            model = AlbertForSequenceClassificationWithPabee.from_pretrained('albert-base-v2')
-            input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
-            labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
-            outputs = model(input_ids, labels=labels)
-            loss, logits = outputs[:2]
+                tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
+                model = AlbertForSequenceClassificationWithPabee.from_pretrained('albert-base-v2')
+                input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0)  # Batch size 1
+                labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
+                outputs = model(input_ids, labels=labels)
+                loss, logits = outputs[:2]
 
         """
 

examples/bert-loses-patience/pabee/modeling_pabee_bert.py

@@ -108,30 +108,30 @@ class BertModelWithPabee(BertModel):
         regression=False,
     ):
         r"""
-    Return:
-        :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
-        last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
-            Sequence of hidden-states at the output of the last layer of the model.
-        pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(batch_size, hidden_size)`):
-            Last layer hidden-state of the first token of the sequence (classification token)
-            further processed by a Linear layer and a Tanh activation function. The Linear
-            layer weights are trained from the next sentence prediction (classification)
-            objective during pre-training.
+        Return:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
+            last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
+                Sequence of hidden-states at the output of the last layer of the model.
+            pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(batch_size, hidden_size)`):
+                Last layer hidden-state of the first token of the sequence (classification token)
+                further processed by a Linear layer and a Tanh activation function. The Linear
+                layer weights are trained from the next sentence prediction (classification)
+                objective during pre-training.
 
-            This output is usually *not* a good summary
-            of the semantic content of the input, you're often better with averaging or pooling
-            the sequence of hidden-states for the whole input sequence.
-        hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
-            of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+                This output is usually *not* a good summary
+                of the semantic content of the input, you're often better with averaging or pooling
+                the sequence of hidden-states for the whole input sequence.
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
 
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
-            :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
 
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
-            heads.
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
         """
 
         if input_ids is not None and inputs_embeds is not None:
@@ -266,44 +266,44 @@ class BertForSequenceClassificationWithPabee(BertPreTrainedModel):
         labels=None,
     ):
         r"""
-        labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
-            Labels for computing the sequence classification/regression loss.
-            Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
-            If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
-            If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+            labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
+                Labels for computing the sequence classification/regression loss.
+                Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
+                If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
+                If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
 
-    Returns:
-        :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
-        loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`label` is provided):
-            Classification (or regression if config.num_labels==1) loss.
-        logits (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, config.num_labels)`):
-            Classification (or regression if config.num_labels==1) scores (before SoftMax).
-        hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
-            of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+        Returns:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
+            loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`label` is provided):
+                Classification (or regression if config.num_labels==1) loss.
+            logits (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, config.num_labels)`):
+                Classification (or regression if config.num_labels==1) scores (before SoftMax).
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
 
-            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-        attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
-            Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
-            :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
 
-            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
-            heads.
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
 
-    Examples::
+        Examples::
 
-        from transformers import BertTokenizer, BertForSequenceClassification
-        from pabee import BertForSequenceClassificationWithPabee
-        import torch
+            from transformers import BertTokenizer, BertForSequenceClassification
+            from pabee import BertForSequenceClassificationWithPabee
+            import torch
 
-        tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
-        model = BertForSequenceClassificationWithPabee.from_pretrained('bert-base-uncased')
+            tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+            model = BertForSequenceClassificationWithPabee.from_pretrained('bert-base-uncased')
 
-        input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
-        labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
-        outputs = model(input_ids, labels=labels)
+            input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
+            labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
+            outputs = model(input_ids, labels=labels)
 
-        loss, logits = outputs[:2]
+            loss, logits = outputs[:2]
 
         """
 

examples/bert-loses-patience/run_glue_with_pabee.py

@@ -120,7 +120,10 @@ def train(args, train_dataset, model, tokenizer):
     # Distributed training (should be after apex fp16 initialization)
     if args.local_rank != -1:
         model = torch.nn.parallel.DistributedDataParallel(
-            model, device_ids=[args.local_rank], output_device=args.local_rank, find_unused_parameters=True,
+            model,
+            device_ids=[args.local_rank],
+            output_device=args.local_rank,
+            find_unused_parameters=True,
         )
 
     # Train!
@@ -151,13 +154,17 @@ def train(args, train_dataset, model, tokenizer):
         logger.info("  Continuing training from epoch %d", epochs_trained)
         logger.info("  Continuing training from global step %d", global_step)
         logger.info(
-            "  Will skip the first %d steps in the first epoch", steps_trained_in_current_epoch,
+            "  Will skip the first %d steps in the first epoch",
+            steps_trained_in_current_epoch,
         )
 
     tr_loss, logging_loss = 0.0, 0.0
     model.zero_grad()
     train_iterator = trange(
-        epochs_trained, int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0],
+        epochs_trained,
+        int(args.num_train_epochs),
+        desc="Epoch",
+        disable=args.local_rank not in [-1, 0],
     )
     set_seed(args)  # Added here for reproductibility
     for _ in train_iterator:
@@ -372,7 +379,11 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
             processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
         )
         features = convert_examples_to_features(
-            examples, tokenizer, label_list=label_list, max_length=args.max_seq_length, output_mode=output_mode,
+            examples,
+            tokenizer,
+            label_list=label_list,
+            max_length=args.max_seq_length,
+            output_mode=output_mode,
         )
         if args.local_rank in [-1, 0]:
             logger.info("Saving features into cached file %s", cached_features_file)
@@ -434,15 +445,24 @@ def main():
         help="The output directory where the model predictions and checkpoints will be written.",
     )
     parser.add_argument(
-        "--patience", default="0", type=str, required=False,
+        "--patience",
+        default="0",
+        type=str,
+        required=False,
     )
     parser.add_argument(
-        "--regression_threshold", default=0, type=float, required=False,
+        "--regression_threshold",
+        default=0,
+        type=float,
+        required=False,
     )
 
     # Other parameters
     parser.add_argument(
-        "--config_name", default="", type=str, help="Pretrained config name or path if not the same as model_name",
+        "--config_name",
+        default="",
+        type=str,
+        help="Pretrained config name or path if not the same as model_name",
     )
     parser.add_argument(
         "--tokenizer_name",
@@ -466,17 +486,27 @@ def main():
     parser.add_argument("--do_train", action="store_true", help="Whether to run training.")
     parser.add_argument("--do_eval", action="store_true", help="Whether to run eval on the dev set.")
     parser.add_argument(
-        "--evaluate_during_training", action="store_true", help="Run evaluation during training at each logging step.",
+        "--evaluate_during_training",
+        action="store_true",
+        help="Run evaluation during training at each logging step.",
     )
     parser.add_argument(
-        "--do_lower_case", action="store_true", help="Set this flag if you are using an uncased model.",
+        "--do_lower_case",
+        action="store_true",
+        help="Set this flag if you are using an uncased model.",
     )
 
     parser.add_argument(
-        "--per_gpu_train_batch_size", default=8, type=int, help="Batch size per GPU/CPU for training.",
+        "--per_gpu_train_batch_size",
+        default=8,
+        type=int,
+        help="Batch size per GPU/CPU for training.",
     )
     parser.add_argument(
-        "--per_gpu_eval_batch_size", default=1, type=int, help="Batch size per GPU/CPU for evaluation.",
+        "--per_gpu_eval_batch_size",
+        default=1,
+        type=int,
+        help="Batch size per GPU/CPU for evaluation.",
     )
     parser.add_argument(
         "--gradient_accumulation_steps",
@@ -485,13 +515,19 @@ def main():
         help="Number of updates steps to accumulate before performing a backward/update pass.",
     )
     parser.add_argument(
-        "--learning_rate", default=5e-5, type=float, help="The initial learning rate for Adam.",
+        "--learning_rate",
+        default=5e-5,
+        type=float,
+        help="The initial learning rate for Adam.",
     )
     parser.add_argument("--weight_decay", default=0.0, type=float, help="Weight decay if we apply some.")
     parser.add_argument("--adam_epsilon", default=1e-8, type=float, help="Epsilon for Adam optimizer.")
     parser.add_argument("--max_grad_norm", default=1.0, type=float, help="Max gradient norm.")
     parser.add_argument(
-        "--num_train_epochs", default=3.0, type=float, help="Total number of training epochs to perform.",
+        "--num_train_epochs",
+        default=3.0,
+        type=float,
+        help="Total number of training epochs to perform.",
     )
     parser.add_argument(
         "--max_steps",
@@ -503,7 +539,10 @@ def main():
 
     parser.add_argument("--logging_steps", type=int, default=500, help="Log every X updates steps.")
     parser.add_argument(
-        "--save_steps", type=int, default=500, help="Save checkpoint every X updates steps.",
+        "--save_steps",
+        type=int,
+        default=500,
+        help="Save checkpoint every X updates steps.",
     )
     parser.add_argument(
         "--eval_all_checkpoints",
@@ -512,10 +551,14 @@ def main():
     )
     parser.add_argument("--no_cuda", action="store_true", help="Avoid using CUDA when available")
     parser.add_argument(
-        "--overwrite_output_dir", action="store_true", help="Overwrite the content of the output directory",
+        "--overwrite_output_dir",
+        action="store_true",
+        help="Overwrite the content of the output directory",
     )
     parser.add_argument(
-        "--overwrite_cache", action="store_true", help="Overwrite the cached training and evaluation sets",
+        "--overwrite_cache",
+        action="store_true",
+        help="Overwrite the cached training and evaluation sets",
     )
     parser.add_argument("--seed", type=int, default=42, help="random seed for initialization")
 
@@ -532,7 +575,10 @@ def main():
         "See details at https://nvidia.github.io/apex/amp.html",
     )
     parser.add_argument(
-        "--local_rank", type=int, default=-1, help="For distributed training: local_rank",
+        "--local_rank",
+        type=int,
+        default=-1,
+        help="For distributed training: local_rank",
     )
     parser.add_argument("--server_ip", type=str, default="", help="For distant debugging.")
     parser.add_argument("--server_port", type=str, default="", help="For distant debugging.")
@@ -634,7 +680,8 @@ def main():
     print("Output Layers Parameters:", output_layers_param_num)
     single_output_layer_param_num = sum(param.numel() for param in model.classifiers[0].parameters())
     print(
-        "Added Output Layers Parameters:", output_layers_param_num - single_output_layer_param_num,
+        "Added Output Layers Parameters:",
+        output_layers_param_num - single_output_layer_param_num,
     )
 
     logger.info("Training/evaluation parameters %s", args)

examples/bert-loses-patience/test_run_glue_with_pabee.py

@@ -1,10 +1,10 @@
 import argparse
 import logging
 import sys
-import unittest
 from unittest.mock import patch
 
 import run_glue_with_pabee
+from transformers.testing_utils import TestCasePlus
 
 
 logging.basicConfig(level=logging.DEBUG)
@@ -19,29 +19,31 @@ def get_setup_file():
     return args.f
 
 
-class PabeeTests(unittest.TestCase):
+class PabeeTests(TestCasePlus):
     def test_run_glue(self):
         stream_handler = logging.StreamHandler(sys.stdout)
         logger.addHandler(stream_handler)
 
-        testargs = """
+        tmp_dir = self.get_auto_remove_tmp_dir()
+        testargs = f"""
             run_glue_with_pabee.py
             --model_type albert
             --model_name_or_path albert-base-v2
             --data_dir ./tests/fixtures/tests_samples/MRPC/
+            --output_dir {tmp_dir}
+            --overwrite_output_dir
             --task_name mrpc
             --do_train
             --do_eval
-            --output_dir ./tests/fixtures/tests_samples/temp_dir
             --per_gpu_train_batch_size=2
             --per_gpu_eval_batch_size=1
             --learning_rate=2e-5
             --max_steps=50
             --warmup_steps=2
-            --overwrite_output_dir
             --seed=42
             --max_seq_length=128
             """.split()
+
         with patch.object(sys, "argv", testargs):
             result = run_glue_with_pabee.main()
             for value in result.values():

examples/bertology/run_bertology.py

@@ -66,9 +66,9 @@ def print_2d_tensor(tensor):
 def compute_heads_importance(
     args, model, eval_dataloader, compute_entropy=True, compute_importance=True, head_mask=None, actually_pruned=False
 ):
-    """ This method shows how to compute:
-        - head attention entropy
-        - head importance scores according to http://arxiv.org/abs/1905.10650
+    """This method shows how to compute:
+    - head attention entropy
+    - head importance scores according to http://arxiv.org/abs/1905.10650
     """
     # Prepare our tensors
     n_layers, n_heads = model.config.num_hidden_layers, model.config.num_attention_heads
@@ -150,8 +150,8 @@ def compute_heads_importance(
 
 
 def mask_heads(args, model, eval_dataloader):
-    """ This method shows how to mask head (set some heads to zero), to test the effect on the network,
-        based on the head importance scores, as described in Michel et al. (http://arxiv.org/abs/1905.10650)
+    """This method shows how to mask head (set some heads to zero), to test the effect on the network,
+    based on the head importance scores, as described in Michel et al. (http://arxiv.org/abs/1905.10650)
     """
     _, head_importance, preds, labels = compute_heads_importance(args, model, eval_dataloader, compute_entropy=False)
     preds = np.argmax(preds, axis=1) if args.output_mode == "classification" else np.squeeze(preds)
@@ -201,8 +201,8 @@ def mask_heads(args, model, eval_dataloader):
 
 
 def prune_heads(args, model, eval_dataloader, head_mask):
-    """ This method shows how to prune head (remove heads weights) based on
-        the head importance scores as described in Michel et al. (http://arxiv.org/abs/1905.10650)
+    """This method shows how to prune head (remove heads weights) based on
+    the head importance scores as described in Michel et al. (http://arxiv.org/abs/1905.10650)
     """
     # Try pruning and test time speedup
     # Pruning is like masking but we actually remove the masked weights
@@ -395,7 +395,8 @@ def main():
         cache_dir=args.cache_dir,
     )
     tokenizer = AutoTokenizer.from_pretrained(
-        args.tokenizer_name if args.tokenizer_name else args.model_name_or_path, cache_dir=args.cache_dir,
+        args.tokenizer_name if args.tokenizer_name else args.model_name_or_path,
+        cache_dir=args.cache_dir,
     )
     model = AutoModelForSequenceClassification.from_pretrained(
         args.model_name_or_path,

examples/conftest.py

@@ -0,0 +1,11 @@
+# tests directory-specific settings - this file is run automatically
+# by pytest before any tests are run
+
+import sys
+from os.path import abspath, dirname, join
+
+
+# allow having multiple repository checkouts and not needing to remember to rerun
+# 'pip install -e .[dev]' when switching between checkouts and running tests.
+git_repo_path = abspath(join(dirname(dirname(__file__)), "src"))
+sys.path.insert(1, git_repo_path)

examples/contrib/mm-imdb/utils_mmimdb.py

@@ -138,6 +138,9 @@ def get_image_transforms():
             transforms.Resize(256),
             transforms.CenterCrop(224),
             transforms.ToTensor(),
-            transforms.Normalize(mean=[0.46777044, 0.44531429, 0.40661017], std=[0.12221994, 0.12145835, 0.14380469],),
+            transforms.Normalize(
+                mean=[0.46777044, 0.44531429, 0.40661017],
+                std=[0.12221994, 0.12145835, 0.14380469],
+            ),
         ]
     )

examples/contrib/run_camembert.py

@@ -30,7 +30,11 @@ def fill_mask(masked_input, model, tokenizer, topk=5):
             )
         else:
             topk_filled_outputs.append(
-                (masked_input.replace(masked_token, predicted_token), values[index].item(), predicted_token,)
+                (
+                    masked_input.replace(masked_token, predicted_token),
+                    values[index].item(),
+                    predicted_token,
+                )
             )
     return topk_filled_outputs
 

examples/contrib/run_openai_gpt.py

@@ -71,10 +71,10 @@ def load_rocstories_dataset(dataset_path):
 
 
 def pre_process_datasets(encoded_datasets, input_len, cap_length, start_token, delimiter_token, clf_token):
-    """ Pre-process datasets containing lists of tuples(story, 1st continuation, 2nd continuation, label)
+    """Pre-process datasets containing lists of tuples(story, 1st continuation, 2nd continuation, label)
 
-        To Transformer inputs of shape (n_batch, n_alternative, length) comprising for each batch, continuation:
-        input_ids[batch, alternative, :] = [start_token] + story[:cap_length] + [delimiter_token] + cont1[:cap_length] + [clf_token]
+    To Transformer inputs of shape (n_batch, n_alternative, length) comprising for each batch, continuation:
+    input_ids[batch, alternative, :] = [start_token] + story[:cap_length] + [delimiter_token] + cont1[:cap_length] + [clf_token]
     """
     tensor_datasets = []
     for dataset in encoded_datasets:
@@ -83,7 +83,10 @@ def pre_process_datasets(encoded_datasets, input_len, cap_length, start_token, d
         mc_token_ids = np.zeros((n_batch, 2), dtype=np.int64)
         lm_labels = np.full((n_batch, 2, input_len), fill_value=-100, dtype=np.int64)
         mc_labels = np.zeros((n_batch,), dtype=np.int64)
-        for i, (story, cont1, cont2, mc_label), in enumerate(dataset):
+        for (
+            i,
+            (story, cont1, cont2, mc_label),
+        ) in enumerate(dataset):
             with_cont1 = [start_token] + story[:cap_length] + [delimiter_token] + cont1[:cap_length] + [clf_token]
             with_cont2 = [start_token] + story[:cap_length] + [delimiter_token] + cont2[:cap_length] + [clf_token]
             input_ids[i, 0, : len(with_cont1)] = with_cont1

examples/contrib/run_swag.py

@@ -629,7 +629,9 @@ def main():
         torch.distributed.barrier()  # Make sure only the first process in distributed training will download model & vocab
 
     config = AutoConfig.from_pretrained(args.config_name if args.config_name else args.model_name_or_path)
-    tokenizer = AutoTokenizer.from_pretrained(args.tokenizer_name if args.tokenizer_name else args.model_name_or_path,)
+    tokenizer = AutoTokenizer.from_pretrained(
+        args.tokenizer_name if args.tokenizer_name else args.model_name_or_path,
+    )
     model = AutoModelForMultipleChoice.from_pretrained(
         args.model_name_or_path, from_tf=bool(".ckpt" in args.model_name_or_path), config=config
     )

examples/deebert/README.md

@@ -0,0 +1,54 @@
+# DeeBERT: Early Exiting for *BERT
+
+This is the code base for the paper [DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference](https://www.aclweb.org/anthology/2020.acl-main.204/), modified from its [original code base](https://github.com/castorini/deebert).
+
+The original code base also has information for downloading sample models that we have trained in advance.
+
+## Usage
+
+There are three scripts in the folder which can be run directly.
+
+In each script, there are several things to modify before running:
+
+* `PATH_TO_DATA`: path to the GLUE dataset.
+* `--output_dir`: path for saving fine-tuned models. Default: `./saved_models`.
+* `--plot_data_dir`: path for saving evaluation results. Default: `./results`. Results are printed to stdout and also saved to `npy` files in this directory to facilitate plotting figures and further analyses.
+* `MODEL_TYPE`: bert or roberta
+* `MODEL_SIZE`: base or large
+* `DATASET`: SST-2, MRPC, RTE, QNLI, QQP, or MNLI
+
+#### train_deebert.sh
+
+This is for fine-tuning DeeBERT models.
+
+#### eval_deebert.sh
+
+This is for evaluating each exit layer for fine-tuned DeeBERT models.
+
+#### entropy_eval.sh
+
+This is for evaluating fine-tuned DeeBERT models, given a number of different early exit entropy thresholds.
+
+
+
+## Citation
+
+Please cite our paper if you find the resource useful:
+```
+@inproceedings{xin-etal-2020-deebert,
+    title = "{D}ee{BERT}: Dynamic Early Exiting for Accelerating {BERT} Inference",
+    author = "Xin, Ji  and
+      Tang, Raphael  and
+      Lee, Jaejun  and
+      Yu, Yaoliang  and
+      Lin, Jimmy",
+    booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
+    month = jul,
+    year = "2020",
+    address = "Online",
+    publisher = "Association for Computational Linguistics",
+    url = "https://www.aclweb.org/anthology/2020.acl-main.204",
+    pages = "2246--2251",
+}
+```
+

examples/deebert/entropy_eval.sh

@@ -0,0 +1,33 @@
+#!/bin/bash
+export CUDA_VISIBLE_DEVICES=0
+
+PATH_TO_DATA=/h/xinji/projects/GLUE
+
+MODEL_TYPE=bert  # bert or roberta
+MODEL_SIZE=base  # base or large
+DATASET=MRPC  # SST-2, MRPC, RTE, QNLI, QQP, or MNLI
+
+MODEL_NAME=${MODEL_TYPE}-${MODEL_SIZE}
+if [ $MODEL_TYPE = 'bert' ]
+then
+  MODEL_NAME=${MODEL_NAME}-uncased
+fi
+
+ENTROPIES="0 0.1 0.2 0.3 0.4 0.5 0.6 0.7"
+
+for ENTROPY in $ENTROPIES; do
+  python -u run_glue_deebert.py \
+    --model_type $MODEL_TYPE \
+    --model_name_or_path ./saved_models/${MODEL_TYPE}-${MODEL_SIZE}/$DATASET/two_stage \
+    --task_name $DATASET \
+    --do_eval \
+    --do_lower_case \
+    --data_dir $PATH_TO_DATA/$DATASET \
+    --output_dir ./saved_models/${MODEL_TYPE}-${MODEL_SIZE}/$DATASET/two_stage \
+    --plot_data_dir ./results/ \
+    --max_seq_length 128 \
+    --early_exit_entropy $ENTROPY \
+    --eval_highway \
+    --overwrite_cache \
+    --per_gpu_eval_batch_size=1
+done

examples/deebert/eval_deebert.sh

@@ -0,0 +1,30 @@
+#!/bin/bash
+export CUDA_VISIBLE_DEVICES=0
+
+PATH_TO_DATA=/h/xinji/projects/GLUE
+
+MODEL_TYPE=bert  # bert or roberta
+MODEL_SIZE=base  # base or large
+DATASET=MRPC  # SST-2, MRPC, RTE, QNLI, QQP, or MNLI
+
+MODEL_NAME=${MODEL_TYPE}-${MODEL_SIZE}
+if [ $MODEL_TYPE = 'bert' ]
+then
+  MODEL_NAME=${MODEL_NAME}-uncased
+fi
+
+
+python -u run_glue_deebert.py  \
+  --model_type $MODEL_TYPE \
+  --model_name_or_path ./saved_models/${MODEL_TYPE}-${MODEL_SIZE}/$DATASET/two_stage \
+  --task_name $DATASET \
+  --do_eval \
+  --do_lower_case \
+  --data_dir $PATH_TO_DATA/$DATASET \
+  --output_dir ./saved_models/${MODEL_TYPE}-${MODEL_SIZE}/$DATASET/two_stage \
+  --plot_data_dir ./results/ \
+  --max_seq_length 128 \
+  --eval_each_highway \
+  --eval_highway \
+  --overwrite_cache \
+  --per_gpu_eval_batch_size=1

examples/deebert/run_glue_deebert.py

@@ -0,0 +1,724 @@
+from __future__ import absolute_import, division, print_function
+
+import argparse
+import glob
+import logging
+import os
+import random
+import time
+
+import numpy as np
+import torch
+from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, TensorDataset
+from torch.utils.data.distributed import DistributedSampler
+from tqdm import tqdm, trange
+
+from src.modeling_highway_bert import DeeBertForSequenceClassification
+from src.modeling_highway_roberta import DeeRobertaForSequenceClassification
+from transformers import (
+    WEIGHTS_NAME,
+    AdamW,
+    BertConfig,
+    BertTokenizer,
+    RobertaConfig,
+    RobertaTokenizer,
+    get_linear_schedule_with_warmup,
+)
+from transformers import glue_compute_metrics as compute_metrics
+from transformers import glue_convert_examples_to_features as convert_examples_to_features
+from transformers import glue_output_modes as output_modes
+from transformers import glue_processors as processors
+
+
+try:
+    from torch.utils.tensorboard import SummaryWriter
+except ImportError:
+    from tensorboardX import SummaryWriter
+
+
+logger = logging.getLogger(__name__)
+
+
+MODEL_CLASSES = {
+    "bert": (BertConfig, DeeBertForSequenceClassification, BertTokenizer),
+    "roberta": (RobertaConfig, DeeRobertaForSequenceClassification, RobertaTokenizer),
+}
+
+
+def set_seed(args):
+    random.seed(args.seed)
+    np.random.seed(args.seed)
+    torch.manual_seed(args.seed)
+    if args.n_gpu > 0:
+        torch.cuda.manual_seed_all(args.seed)
+
+
+def get_wanted_result(result):
+    if "spearmanr" in result:
+        print_result = result["spearmanr"]
+    elif "f1" in result:
+        print_result = result["f1"]
+    elif "mcc" in result:
+        print_result = result["mcc"]
+    elif "acc" in result:
+        print_result = result["acc"]
+    else:
+        raise ValueError("Primary metric unclear in the results")
+    return print_result
+
+
+def train(args, train_dataset, model, tokenizer, train_highway=False):
+    """ Train the model """
+    if args.local_rank in [-1, 0]:
+        tb_writer = SummaryWriter()
+
+    args.train_batch_size = args.per_gpu_train_batch_size * max(1, args.n_gpu)
+    train_sampler = RandomSampler(train_dataset) if args.local_rank == -1 else DistributedSampler(train_dataset)
+    train_dataloader = DataLoader(train_dataset, sampler=train_sampler, batch_size=args.train_batch_size)
+
+    if args.max_steps > 0:
+        t_total = args.max_steps
+        args.num_train_epochs = args.max_steps // (len(train_dataloader) // args.gradient_accumulation_steps) + 1
+    else:
+        t_total = len(train_dataloader) // args.gradient_accumulation_steps * args.num_train_epochs
+
+    # Prepare optimizer and schedule (linear warmup and decay)
+    no_decay = ["bias", "LayerNorm.weight"]
+    if train_highway:
+        optimizer_grouped_parameters = [
+            {
+                "params": [
+                    p
+                    for n, p in model.named_parameters()
+                    if ("highway" in n) and (not any(nd in n for nd in no_decay))
+                ],
+                "weight_decay": args.weight_decay,
+            },
+            {
+                "params": [
+                    p for n, p in model.named_parameters() if ("highway" in n) and (any(nd in n for nd in no_decay))
+                ],
+                "weight_decay": 0.0,
+            },
+        ]
+    else:
+        optimizer_grouped_parameters = [
+            {
+                "params": [
+                    p
+                    for n, p in model.named_parameters()
+                    if ("highway" not in n) and (not any(nd in n for nd in no_decay))
+                ],
+                "weight_decay": args.weight_decay,
+            },
+            {
+                "params": [
+                    p
+                    for n, p in model.named_parameters()
+                    if ("highway" not in n) and (any(nd in n for nd in no_decay))
+                ],
+                "weight_decay": 0.0,
+            },
+        ]
+    optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
+    scheduler = get_linear_schedule_with_warmup(
+        optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total
+    )
+    if args.fp16:
+        try:
+            from apex import amp
+        except ImportError:
+            raise ImportError("Please install apex from https://www.github.com/nvidia/apex to use fp16 training.")
+        model, optimizer = amp.initialize(model, optimizer, opt_level=args.fp16_opt_level)
+
+    # multi-gpu training (should be after apex fp16 initialization)
+    if args.n_gpu > 1:
+        model = torch.nn.DataParallel(model)
+
+    # Distributed training (should be after apex fp16 initialization)
+    if args.local_rank != -1:
+        model = torch.nn.parallel.DistributedDataParallel(
+            model, device_ids=[args.local_rank], output_device=args.local_rank, find_unused_parameters=True
+        )
+
+    # Train!
+    logger.info("***** Running training *****")
+    logger.info("  Num examples = %d", len(train_dataset))
+    logger.info("  Num Epochs = %d", args.num_train_epochs)
+    logger.info("  Instantaneous batch size per GPU = %d", args.per_gpu_train_batch_size)
+    logger.info(
+        "  Total train batch size (w. parallel, distributed & accumulation) = %d",
+        args.train_batch_size
+        * args.gradient_accumulation_steps
+        * (torch.distributed.get_world_size() if args.local_rank != -1 else 1),
+    )
+    logger.info("  Gradient Accumulation steps = %d", args.gradient_accumulation_steps)
+    logger.info("  Total optimization steps = %d", t_total)
+
+    global_step = 0
+    tr_loss, logging_loss = 0.0, 0.0
+    model.zero_grad()
+    train_iterator = trange(int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0])
+    set_seed(args)  # Added here for reproductibility (even between python 2 and 3)
+    for _ in train_iterator:
+        epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
+        for step, batch in enumerate(epoch_iterator):
+            model.train()
+            batch = tuple(t.to(args.device) for t in batch)
+            inputs = {"input_ids": batch[0], "attention_mask": batch[1], "labels": batch[3]}
+            if args.model_type != "distilbert":
+                inputs["token_type_ids"] = (
+                    batch[2] if args.model_type in ["bert", "xlnet"] else None
+                )  # XLM, DistilBERT and RoBERTa don't use segment_ids
+            inputs["train_highway"] = train_highway
+            outputs = model(**inputs)
+            loss = outputs[0]  # model outputs are always tuple in transformers (see doc)
+
+            if args.n_gpu > 1:
+                loss = loss.mean()  # mean() to average on multi-gpu parallel training
+            if args.gradient_accumulation_steps > 1:
+                loss = loss / args.gradient_accumulation_steps
+
+            if args.fp16:
+                with amp.scale_loss(loss, optimizer) as scaled_loss:
+                    scaled_loss.backward()
+            else:
+                loss.backward()
+
+            tr_loss += loss.item()
+            if (step + 1) % args.gradient_accumulation_steps == 0:
+                if args.fp16:
+                    torch.nn.utils.clip_grad_norm_(amp.master_params(optimizer), args.max_grad_norm)
+                else:
+                    torch.nn.utils.clip_grad_norm_(model.parameters(), args.max_grad_norm)
+
+                optimizer.step()
+                scheduler.step()  # Update learning rate schedule
+                model.zero_grad()
+                global_step += 1
+
+                if args.local_rank in [-1, 0] and args.logging_steps > 0 and global_step % args.logging_steps == 0:
+                    # Log metrics
+                    if (
+                        args.local_rank == -1 and args.evaluate_during_training
+                    ):  # Only evaluate when single GPU otherwise metrics may not average well
+                        results = evaluate(args, model, tokenizer)
+                        for key, value in results.items():
+                            tb_writer.add_scalar("eval_{}".format(key), value, global_step)
+                    tb_writer.add_scalar("lr", scheduler.get_lr()[0], global_step)
+                    tb_writer.add_scalar("loss", (tr_loss - logging_loss) / args.logging_steps, global_step)
+                    logging_loss = tr_loss
+
+                if args.local_rank in [-1, 0] and args.save_steps > 0 and global_step % args.save_steps == 0:
+                    # Save model checkpoint
+                    output_dir = os.path.join(args.output_dir, "checkpoint-{}".format(global_step))
+                    if not os.path.exists(output_dir):
+                        os.makedirs(output_dir)
+                    model_to_save = (
+                        model.module if hasattr(model, "module") else model
+                    )  # Take care of distributed/parallel training
+                    model_to_save.save_pretrained(output_dir)
+                    torch.save(args, os.path.join(output_dir, "training_args.bin"))
+                    logger.info("Saving model checkpoint to %s", output_dir)
+
+            if args.max_steps > 0 and global_step > args.max_steps:
+                epoch_iterator.close()
+                break
+        if args.max_steps > 0 and global_step > args.max_steps:
+            train_iterator.close()
+            break
+
+    if args.local_rank in [-1, 0]:
+        tb_writer.close()
+
+    return global_step, tr_loss / global_step
+
+
+def evaluate(args, model, tokenizer, prefix="", output_layer=-1, eval_highway=False):
+    # Loop to handle MNLI double evaluation (matched, mis-matched)
+    eval_task_names = ("mnli", "mnli-mm") if args.task_name == "mnli" else (args.task_name,)
+    eval_outputs_dirs = (args.output_dir, args.output_dir + "-MM") if args.task_name == "mnli" else (args.output_dir,)
+
+    results = {}
+    for eval_task, eval_output_dir in zip(eval_task_names, eval_outputs_dirs):
+        eval_dataset = load_and_cache_examples(args, eval_task, tokenizer, evaluate=True)
+
+        if not os.path.exists(eval_output_dir) and args.local_rank in [-1, 0]:
+            os.makedirs(eval_output_dir)
+
+        args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
+        # Note that DistributedSampler samples randomly
+        eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
+        eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
+
+        # multi-gpu eval
+        if args.n_gpu > 1:
+            model = torch.nn.DataParallel(model)
+
+        # Eval!
+        logger.info("***** Running evaluation {} *****".format(prefix))
+        logger.info("  Num examples = %d", len(eval_dataset))
+        logger.info("  Batch size = %d", args.eval_batch_size)
+        eval_loss = 0.0
+        nb_eval_steps = 0
+        preds = None
+        out_label_ids = None
+        exit_layer_counter = {(i + 1): 0 for i in range(model.num_layers)}
+        st = time.time()
+        for batch in tqdm(eval_dataloader, desc="Evaluating"):
+            model.eval()
+            batch = tuple(t.to(args.device) for t in batch)
+
+            with torch.no_grad():
+                inputs = {"input_ids": batch[0], "attention_mask": batch[1], "labels": batch[3]}
+                if args.model_type != "distilbert":
+                    inputs["token_type_ids"] = (
+                        batch[2] if args.model_type in ["bert", "xlnet"] else None
+                    )  # XLM, DistilBERT and RoBERTa don't use segment_ids
+                if output_layer >= 0:
+                    inputs["output_layer"] = output_layer
+                outputs = model(**inputs)
+                if eval_highway:
+                    exit_layer_counter[outputs[-1]] += 1
+                tmp_eval_loss, logits = outputs[:2]
+
+                eval_loss += tmp_eval_loss.mean().item()
+            nb_eval_steps += 1
+            if preds is None:
+                preds = logits.detach().cpu().numpy()
+                out_label_ids = inputs["labels"].detach().cpu().numpy()
+            else:
+                preds = np.append(preds, logits.detach().cpu().numpy(), axis=0)
+                out_label_ids = np.append(out_label_ids, inputs["labels"].detach().cpu().numpy(), axis=0)
+        eval_time = time.time() - st
+        logger.info("Eval time: {}".format(eval_time))
+
+        eval_loss = eval_loss / nb_eval_steps
+        if args.output_mode == "classification":
+            preds = np.argmax(preds, axis=1)
+        elif args.output_mode == "regression":
+            preds = np.squeeze(preds)
+        result = compute_metrics(eval_task, preds, out_label_ids)
+        results.update(result)
+
+        if eval_highway:
+            logger.info("Exit layer counter: {}".format(exit_layer_counter))
+            actual_cost = sum([l * c for l, c in exit_layer_counter.items()])
+            full_cost = len(eval_dataloader) * model.num_layers
+            logger.info("Expected saving: {}".format(actual_cost / full_cost))
+            if args.early_exit_entropy >= 0:
+                save_fname = (
+                    args.plot_data_dir
+                    + "/"
+                    + args.model_name_or_path[2:]
+                    + "/entropy_{}.npy".format(args.early_exit_entropy)
+                )
+                if not os.path.exists(os.path.dirname(save_fname)):
+                    os.makedirs(os.path.dirname(save_fname))
+                print_result = get_wanted_result(result)
+                np.save(save_fname, np.array([exit_layer_counter, eval_time, actual_cost / full_cost, print_result]))
+                logger.info("Entropy={}\tResult={:.2f}".format(args.early_exit_entropy, 100 * print_result))
+
+        output_eval_file = os.path.join(eval_output_dir, prefix, "eval_results.txt")
+        with open(output_eval_file, "w") as writer:
+            logger.info("***** Eval results {} *****".format(prefix))
+            for key in sorted(result.keys()):
+                logger.info("  %s = %s", key, str(result[key]))
+                writer.write("%s = %s\n" % (key, str(result[key])))
+
+    return results
+
+
+def load_and_cache_examples(args, task, tokenizer, evaluate=False):
+    if args.local_rank not in [-1, 0] and not evaluate:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
+
+    processor = processors[task]()
+    output_mode = output_modes[task]
+    # Load data features from cache or dataset file
+    cached_features_file = os.path.join(
+        args.data_dir,
+        "cached_{}_{}_{}_{}".format(
+            "dev" if evaluate else "train",
+            list(filter(None, args.model_name_or_path.split("/"))).pop(),
+            str(args.max_seq_length),
+            str(task),
+        ),
+    )
+    if os.path.exists(cached_features_file) and not args.overwrite_cache:
+        logger.info("Loading features from cached file %s", cached_features_file)
+        features = torch.load(cached_features_file)
+    else:
+        logger.info("Creating features from dataset file at %s", args.data_dir)
+        label_list = processor.get_labels()
+        if task in ["mnli", "mnli-mm"] and args.model_type in ["roberta"]:
+            # HACK(label indices are swapped in RoBERTa pretrained model)
+            label_list[1], label_list[2] = label_list[2], label_list[1]
+        examples = (
+            processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
+        )
+        features = convert_examples_to_features(
+            examples,
+            tokenizer,
+            label_list=label_list,
+            max_length=args.max_seq_length,
+            output_mode=output_mode,
+        )
+        if args.local_rank in [-1, 0]:
+            logger.info("Saving features into cached file %s", cached_features_file)
+            torch.save(features, cached_features_file)
+
+    if args.local_rank == 0 and not evaluate:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training process the dataset, and the others will use the cache
+
+    # Convert to Tensors and build dataset
+    all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
+    all_attention_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
+
+    if features[0].token_type_ids is None:
+        # For RoBERTa (a potential bug!)
+        all_token_type_ids = torch.tensor([[0] * args.max_seq_length for f in features], dtype=torch.long)
+    else:
+        all_token_type_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)
+    if output_mode == "classification":
+        all_labels = torch.tensor([f.label for f in features], dtype=torch.long)
+    elif output_mode == "regression":
+        all_labels = torch.tensor([f.label for f in features], dtype=torch.float)
+
+    dataset = TensorDataset(all_input_ids, all_attention_mask, all_token_type_ids, all_labels)
+    return dataset
+
+
+def main():
+    parser = argparse.ArgumentParser()
+
+    # Required parameters
+    parser.add_argument(
+        "--data_dir",
+        default=None,
+        type=str,
+        required=True,
+        help="The input data dir. Should contain the .tsv files (or other data files) for the task.",
+    )
+    parser.add_argument(
+        "--model_type",
+        default=None,
+        type=str,
+        required=True,
+        help="Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()),
+    )
+    parser.add_argument(
+        "--model_name_or_path",
+        default=None,
+        type=str,
+        required=True,
+        help="Path to pre-trained model or shortcut name.",
+    )
+    parser.add_argument(
+        "--task_name",
+        default=None,
+        type=str,
+        required=True,
+        help="The name of the task to train selected in the list: " + ", ".join(processors.keys()),
+    )
+    parser.add_argument(
+        "--output_dir",
+        default=None,
+        type=str,
+        required=True,
+        help="The output directory where the model predictions and checkpoints will be written.",
+    )
+    parser.add_argument(
+        "--plot_data_dir",
+        default="./plotting/",
+        type=str,
+        required=False,
+        help="The directory to store data for plotting figures.",
+    )
+
+    # Other parameters
+    parser.add_argument(
+        "--config_name", default="", type=str, help="Pretrained config name or path if not the same as model_name"
+    )
+    parser.add_argument(
+        "--tokenizer_name",
+        default="",
+        type=str,
+        help="Pretrained tokenizer name or path if not the same as model_name",
+    )
+    parser.add_argument(
+        "--cache_dir",
+        default="",
+        type=str,
+        help="Where do you want to store the pre-trained models downloaded from s3",
+    )
+    parser.add_argument(
+        "--max_seq_length",
+        default=128,
+        type=int,
+        help="The maximum total input sequence length after tokenization. Sequences longer "
+        "than this will be truncated, sequences shorter will be padded.",
+    )
+    parser.add_argument("--do_train", action="store_true", help="Whether to run training.")
+    parser.add_argument("--do_eval", action="store_true", help="Whether to run eval on the dev set.")
+    parser.add_argument(
+        "--evaluate_during_training", action="store_true", help="Rul evaluation during training at each logging step."
+    )
+    parser.add_argument(
+        "--do_lower_case", action="store_true", help="Set this flag if you are using an uncased model."
+    )
+    parser.add_argument("--eval_each_highway", action="store_true", help="Set this flag to evaluate each highway.")
+    parser.add_argument(
+        "--eval_after_first_stage",
+        action="store_true",
+        help="Set this flag to evaluate after training only bert (not highway).",
+    )
+    parser.add_argument("--eval_highway", action="store_true", help="Set this flag if it's evaluating highway models")
+
+    parser.add_argument("--per_gpu_train_batch_size", default=8, type=int, help="Batch size per GPU/CPU for training.")
+    parser.add_argument(
+        "--per_gpu_eval_batch_size", default=8, type=int, help="Batch size per GPU/CPU for evaluation."
+    )
+    parser.add_argument(
+        "--gradient_accumulation_steps",
+        type=int,
+        default=1,
+        help="Number of updates steps to accumulate before performing a backward/update pass.",
+    )
+    parser.add_argument("--learning_rate", default=5e-5, type=float, help="The initial learning rate for Adam.")
+    parser.add_argument("--weight_decay", default=0.0, type=float, help="Weight deay if we apply some.")
+    parser.add_argument("--adam_epsilon", default=1e-8, type=float, help="Epsilon for Adam optimizer.")
+    parser.add_argument("--max_grad_norm", default=1.0, type=float, help="Max gradient norm.")
+    parser.add_argument(
+        "--num_train_epochs", default=3.0, type=float, help="Total number of training epochs to perform."
+    )
+    parser.add_argument(
+        "--max_steps",
+        default=-1,
+        type=int,
+        help="If > 0: set total number of training steps to perform. Override num_train_epochs.",
+    )
+    parser.add_argument("--warmup_steps", default=0, type=int, help="Linear warmup over warmup_steps.")
+    parser.add_argument("--early_exit_entropy", default=-1, type=float, help="Entropy threshold for early exit.")
+
+    parser.add_argument("--logging_steps", type=int, default=50, help="Log every X updates steps.")
+    parser.add_argument("--save_steps", type=int, default=50, help="Save checkpoint every X updates steps.")
+    parser.add_argument(
+        "--eval_all_checkpoints",
+        action="store_true",
+        help="Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number",
+    )
+    parser.add_argument("--no_cuda", action="store_true", help="Avoid using CUDA when available")
+    parser.add_argument(
+        "--overwrite_output_dir", action="store_true", help="Overwrite the content of the output directory"
+    )
+    parser.add_argument(
+        "--overwrite_cache", action="store_true", help="Overwrite the cached training and evaluation sets"
+    )
+    parser.add_argument("--seed", type=int, default=42, help="random seed for initialization")
+
+    parser.add_argument(
+        "--fp16",
+        action="store_true",
+        help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit",
+    )
+    parser.add_argument(
+        "--fp16_opt_level",
+        type=str,
+        default="O1",
+        help="For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+        "See details at https://nvidia.github.io/apex/amp.html",
+    )
+    parser.add_argument("--local_rank", type=int, default=-1, help="For distributed training: local_rank")
+    parser.add_argument("--server_ip", type=str, default="", help="For distant debugging.")
+    parser.add_argument("--server_port", type=str, default="", help="For distant debugging.")
+    args = parser.parse_args()
+
+    if (
+        os.path.exists(args.output_dir)
+        and os.listdir(args.output_dir)
+        and args.do_train
+        and not args.overwrite_output_dir
+    ):
+        raise ValueError(
+            "Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(
+                args.output_dir
+            )
+        )
+
+    # Setup distant debugging if needed
+    if args.server_ip and args.server_port:
+        # Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
+        import ptvsd
+
+        print("Waiting for debugger attach")
+        ptvsd.enable_attach(address=(args.server_ip, args.server_port), redirect_output=True)
+        ptvsd.wait_for_attach()
+
+    # Setup CUDA, GPU & distributed training
+    if args.local_rank == -1 or args.no_cuda:
+        device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
+        args.n_gpu = torch.cuda.device_count()
+    else:  # Initializes the distributed backend which will take care of sychronizing nodes/GPUs
+        torch.cuda.set_device(args.local_rank)
+        device = torch.device("cuda", args.local_rank)
+        torch.distributed.init_process_group(backend="nccl")
+        args.n_gpu = 1
+    args.device = device
+
+    # Setup logging
+    logging.basicConfig(
+        format="%(asctime)s - %(levelname)s - %(name)s -   %(message)s",
+        datefmt="%m/%d/%Y %H:%M:%S",
+        level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN,
+    )
+    logger.warning(
+        "Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s",
+        args.local_rank,
+        device,
+        args.n_gpu,
+        bool(args.local_rank != -1),
+        args.fp16,
+    )
+
+    # Set seed
+    set_seed(args)
+
+    # Prepare GLUE task
+    args.task_name = args.task_name.lower()
+    if args.task_name not in processors:
+        raise ValueError("Task not found: %s" % (args.task_name))
+    processor = processors[args.task_name]()
+    args.output_mode = output_modes[args.task_name]
+    label_list = processor.get_labels()
+    num_labels = len(label_list)
+
+    # Load pretrained model and tokenizer
+    if args.local_rank not in [-1, 0]:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training will download model & vocab
+
+    args.model_type = args.model_type.lower()
+    config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
+    config = config_class.from_pretrained(
+        args.config_name if args.config_name else args.model_name_or_path,
+        num_labels=num_labels,
+        finetuning_task=args.task_name,
+        cache_dir=args.cache_dir if args.cache_dir else None,
+    )
+    tokenizer = tokenizer_class.from_pretrained(
+        args.tokenizer_name if args.tokenizer_name else args.model_name_or_path,
+        do_lower_case=args.do_lower_case,
+        cache_dir=args.cache_dir if args.cache_dir else None,
+    )
+    model = model_class.from_pretrained(
+        args.model_name_or_path,
+        from_tf=bool(".ckpt" in args.model_name_or_path),
+        config=config,
+        cache_dir=args.cache_dir if args.cache_dir else None,
+    )
+
+    if args.model_type == "bert":
+        model.bert.encoder.set_early_exit_entropy(args.early_exit_entropy)
+        model.bert.init_highway_pooler()
+    elif args.model_type == "roberta":
+        model.roberta.encoder.set_early_exit_entropy(args.early_exit_entropy)
+        model.roberta.init_highway_pooler()
+    else:
+        raise NotImplementedError()
+
+    if args.local_rank == 0:
+        torch.distributed.barrier()  # Make sure only the first process in distributed training will download model & vocab
+
+    model.to(args.device)
+
+    logger.info("Training/evaluation parameters %s", args)
+
+    # Training
+    if args.do_train:
+        train_dataset = load_and_cache_examples(args, args.task_name, tokenizer, evaluate=False)
+        global_step, tr_loss = train(args, train_dataset, model, tokenizer)
+        logger.info(" global_step = %s, average loss = %s", global_step, tr_loss)
+
+        if args.eval_after_first_stage:
+            result = evaluate(args, model, tokenizer, prefix="")
+            print_result = get_wanted_result(result)
+
+        train(args, train_dataset, model, tokenizer, train_highway=True)
+
+    # Saving best-practices: if you use defaults names for the model, you can reload it using from_pretrained()
+    if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0):
+        # Create output directory if needed
+        if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
+            os.makedirs(args.output_dir)
+
+        logger.info("Saving model checkpoint to %s", args.output_dir)
+        # Save a trained model, configuration and tokenizer using `save_pretrained()`.
+        # They can then be reloaded using `from_pretrained()`
+        model_to_save = (
+            model.module if hasattr(model, "module") else model
+        )  # Take care of distributed/parallel training
+        model_to_save.save_pretrained(args.output_dir)
+        tokenizer.save_pretrained(args.output_dir)
+
+        # Good practice: save your training arguments together with the trained model
+        torch.save(args, os.path.join(args.output_dir, "training_args.bin"))
+
+        # Load a trained model and vocabulary that you have fine-tuned
+        model = model_class.from_pretrained(args.output_dir)
+        tokenizer = tokenizer_class.from_pretrained(args.output_dir)
+        model.to(args.device)
+
+    # Evaluation
+    results = {}
+    if args.do_eval and args.local_rank in [-1, 0]:
+        tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
+        checkpoints = [args.output_dir]
+        if args.eval_all_checkpoints:
+            checkpoints = list(
+                os.path.dirname(c) for c in sorted(glob.glob(args.output_dir + "/**/" + WEIGHTS_NAME, recursive=True))
+            )
+            logging.getLogger("transformers.modeling_utils").setLevel(logging.WARN)  # Reduce logging
+        logger.info("Evaluate the following checkpoints: %s", checkpoints)
+        for checkpoint in checkpoints:
+            global_step = checkpoint.split("-")[-1] if len(checkpoints) > 1 else ""
+            prefix = checkpoint.split("/")[-1] if checkpoint.find("checkpoint") != -1 else ""
+
+            model = model_class.from_pretrained(checkpoint)
+            if args.model_type == "bert":
+                model.bert.encoder.set_early_exit_entropy(args.early_exit_entropy)
+            elif args.model_type == "roberta":
+                model.roberta.encoder.set_early_exit_entropy(args.early_exit_entropy)
+            else:
+                raise NotImplementedError()
+
+            model.to(args.device)
+            result = evaluate(args, model, tokenizer, prefix=prefix, eval_highway=args.eval_highway)
+            print_result = get_wanted_result(result)
+            logger.info("Result: {}".format(print_result))
+            if args.eval_each_highway:
+                last_layer_results = print_result
+                each_layer_results = []
+                for i in range(model.num_layers):
+                    logger.info("\n")
+                    _result = evaluate(
+                        args, model, tokenizer, prefix=prefix, output_layer=i, eval_highway=args.eval_highway
+                    )
+                    if i + 1 < model.num_layers:
+                        each_layer_results.append(get_wanted_result(_result))
+                each_layer_results.append(last_layer_results)
+                save_fname = args.plot_data_dir + "/" + args.model_name_or_path[2:] + "/each_layer.npy"
+                if not os.path.exists(os.path.dirname(save_fname)):
+                    os.makedirs(os.path.dirname(save_fname))
+                np.save(save_fname, np.array(each_layer_results))
+                info_str = "Score of each layer:"
+                for i in range(model.num_layers):
+                    info_str += " {:.2f}".format(100 * each_layer_results[i])
+                logger.info(info_str)
+            result = dict((k + "_{}".format(global_step), v) for k, v in result.items())
+            results.update(result)
+
+    return results
+
+
+if __name__ == "__main__":
+    main()

examples/deebert/src/modeling_highway_bert.py

@@ -0,0 +1,396 @@
+import torch
+from torch import nn
+from torch.nn import CrossEntropyLoss, MSELoss
+
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.modeling_bert import (
+    BERT_INPUTS_DOCSTRING,
+    BERT_START_DOCSTRING,
+    BertEmbeddings,
+    BertLayer,
+    BertPooler,
+    BertPreTrainedModel,
+)
+
+
+def entropy(x):
+    """Calculate entropy of a pre-softmax logit Tensor"""
+    exp_x = torch.exp(x)
+    A = torch.sum(exp_x, dim=1)  # sum of exp(x_i)
+    B = torch.sum(x * exp_x, dim=1)  # sum of x_i * exp(x_i)
+    return torch.log(A) - B / A
+
+
+class DeeBertEncoder(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.output_attentions = config.output_attentions
+        self.output_hidden_states = config.output_hidden_states
+        self.layer = nn.ModuleList([BertLayer(config) for _ in range(config.num_hidden_layers)])
+        self.highway = nn.ModuleList([BertHighway(config) for _ in range(config.num_hidden_layers)])
+
+        self.early_exit_entropy = [-1 for _ in range(config.num_hidden_layers)]
+
+    def set_early_exit_entropy(self, x):
+        if (type(x) is float) or (type(x) is int):
+            for i in range(len(self.early_exit_entropy)):
+                self.early_exit_entropy[i] = x
+        else:
+            self.early_exit_entropy = x
+
+    def init_highway_pooler(self, pooler):
+        loaded_model = pooler.state_dict()
+        for highway in self.highway:
+            for name, param in highway.pooler.state_dict().items():
+                param.copy_(loaded_model[name])
+
+    def forward(
+        self,
+        hidden_states,
+        attention_mask=None,
+        head_mask=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+    ):
+        all_hidden_states = ()
+        all_attentions = ()
+        all_highway_exits = ()
+        for i, layer_module in enumerate(self.layer):
+            if self.output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer_outputs = layer_module(
+                hidden_states, attention_mask, head_mask[i], encoder_hidden_states, encoder_attention_mask
+            )
+            hidden_states = layer_outputs[0]
+
+            if self.output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+            current_outputs = (hidden_states,)
+            if self.output_hidden_states:
+                current_outputs = current_outputs + (all_hidden_states,)
+            if self.output_attentions:
+                current_outputs = current_outputs + (all_attentions,)
+
+            highway_exit = self.highway[i](current_outputs)
+            # logits, pooled_output
+
+            if not self.training:
+                highway_logits = highway_exit[0]
+                highway_entropy = entropy(highway_logits)
+                highway_exit = highway_exit + (highway_entropy,)  # logits, hidden_states(?), entropy
+                all_highway_exits = all_highway_exits + (highway_exit,)
+
+                if highway_entropy < self.early_exit_entropy[i]:
+                    new_output = (highway_logits,) + current_outputs[1:] + (all_highway_exits,)
+                    raise HighwayException(new_output, i + 1)
+            else:
+                all_highway_exits = all_highway_exits + (highway_exit,)
+
+        # Add last layer
+        if self.output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        outputs = (hidden_states,)
+        if self.output_hidden_states:
+            outputs = outputs + (all_hidden_states,)
+        if self.output_attentions:
+            outputs = outputs + (all_attentions,)
+
+        outputs = outputs + (all_highway_exits,)
+        return outputs  # last-layer hidden state, (all hidden states), (all attentions), all highway exits
+
+
+@add_start_docstrings(
+    "The Bert Model transformer with early exiting (DeeBERT). ",
+    BERT_START_DOCSTRING,
+)
+class DeeBertModel(BertPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.config = config
+
+        self.embeddings = BertEmbeddings(config)
+        self.encoder = DeeBertEncoder(config)
+        self.pooler = BertPooler(config)
+
+        self.init_weights()
+
+    def init_highway_pooler(self):
+        self.encoder.init_highway_pooler(self.pooler)
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
+    def _prune_heads(self, heads_to_prune):
+        """Prunes heads of the model.
+        heads_to_prune: dict of {layer_num: list of heads to prune in this layer}
+        See base class PreTrainedModel
+        """
+        for layer, heads in heads_to_prune.items():
+            self.encoder.layer[layer].attention.prune_heads(heads)
+
+    @add_start_docstrings_to_callable(BERT_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+    ):
+        r"""
+        Return:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
+            last_hidden_state (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`):
+                Sequence of hidden-states at the output of the last layer of the model.
+            pooler_output (:obj:`torch.FloatTensor`: of shape :obj:`(batch_size, hidden_size)`):
+                Last layer hidden-state of the first token of the sequence (classification token)
+                further processed by a Linear layer and a Tanh activation function. The Linear
+                layer weights are trained from the next sentence prediction (classification)
+                objective during pre-training.
+
+                This output is usually *not* a good summary
+                of the semantic content of the input, you're often better with averaging or pooling
+                the sequence of hidden-states for the whole input sequence.
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``output_hidden_states=True`` is passed or when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``output_attentions=True`` is passed or when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
+            highway_exits (:obj:`tuple(tuple(torch.Tensor))`:
+                Tuple of each early exit's results (total length: number of layers)
+                Each tuple is again, a tuple of length 2 - the first entry is logits and the second entry is hidden states.
+        """
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if attention_mask is None:
+            attention_mask = torch.ones(input_shape, device=device)
+        if encoder_attention_mask is None:
+            encoder_attention_mask = torch.ones(input_shape, device=device)
+        if token_type_ids is None:
+            token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
+
+        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+        # ourselves in which case we just need to make it broadcastable to all heads.
+        extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(attention_mask, input_shape, device)
+
+        # If a 2D ou 3D attention mask is provided for the cross-attention
+        # we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
+        if encoder_attention_mask.dim() == 3:
+            encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
+        if encoder_attention_mask.dim() == 2:
+            encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+
+        encoder_extended_attention_mask = encoder_extended_attention_mask.to(
+            dtype=next(self.parameters()).dtype
+        )  # fp16 compatibility
+        encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+        embedding_output = self.embeddings(
+            input_ids=input_ids, position_ids=position_ids, token_type_ids=token_type_ids, inputs_embeds=inputs_embeds
+        )
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_mask=extended_attention_mask,
+            head_mask=head_mask,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_extended_attention_mask,
+        )
+        sequence_output = encoder_outputs[0]
+        pooled_output = self.pooler(sequence_output)
+
+        outputs = (sequence_output, pooled_output,) + encoder_outputs[
+            1:
+        ]  # add hidden_states and attentions if they are here
+        return outputs  # sequence_output, pooled_output, (hidden_states), (attentions), highway exits
+
+
+class HighwayException(Exception):
+    def __init__(self, message, exit_layer):
+        self.message = message
+        self.exit_layer = exit_layer  # start from 1!
+
+
+class BertHighway(nn.Module):
+    """A module to provide a shortcut
+    from (the output of one non-final BertLayer in BertEncoder) to (cross-entropy computation in BertForSequenceClassification)
+    """
+
+    def __init__(self, config):
+        super().__init__()
+        self.pooler = BertPooler(config)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+    def forward(self, encoder_outputs):
+        # Pooler
+        pooler_input = encoder_outputs[0]
+        pooler_output = self.pooler(pooler_input)
+        # "return" pooler_output
+
+        # BertModel
+        bmodel_output = (pooler_input, pooler_output) + encoder_outputs[1:]
+        # "return" bodel_output
+
+        # Dropout and classification
+        pooled_output = bmodel_output[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        return logits, pooled_output
+
+
+@add_start_docstrings(
+    """Bert Model (with early exiting - DeeBERT) with a classifier on top,
+    also takes care of multi-layer training. """,
+    BERT_START_DOCSTRING,
+)
+class DeeBertForSequenceClassification(BertPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.num_layers = config.num_hidden_layers
+
+        self.bert = DeeBertModel(config)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.classifier = nn.Linear(config.hidden_size, self.config.num_labels)
+
+        self.init_weights()
+
+    @add_start_docstrings_to_callable(BERT_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        labels=None,
+        output_layer=-1,
+        train_highway=False,
+    ):
+        r"""
+            labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
+                Labels for computing the sequence classification/regression loss.
+                Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
+                If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
+                If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+
+        Returns:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.BertConfig`) and inputs:
+            loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`label` is provided):
+                Classification (or regression if config.num_labels==1) loss.
+            logits (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, config.num_labels)`):
+                Classification (or regression if config.num_labels==1) scores (before SoftMax).
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``output_hidden_states=True`` is passed or when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``output_attentions=True`` is passed or when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
+            highway_exits (:obj:`tuple(tuple(torch.Tensor))`:
+                Tuple of each early exit's results (total length: number of layers)
+                Each tuple is again, a tuple of length 2 - the first entry is logits and the second entry is hidden states.
+        """
+
+        exit_layer = self.num_layers
+        try:
+            outputs = self.bert(
+                input_ids,
+                attention_mask=attention_mask,
+                token_type_ids=token_type_ids,
+                position_ids=position_ids,
+                head_mask=head_mask,
+                inputs_embeds=inputs_embeds,
+            )
+            # sequence_output, pooled_output, (hidden_states), (attentions), highway exits
+
+            pooled_output = outputs[1]
+
+            pooled_output = self.dropout(pooled_output)
+            logits = self.classifier(pooled_output)
+            outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+        except HighwayException as e:
+            outputs = e.message
+            exit_layer = e.exit_layer
+            logits = outputs[0]
+
+        if not self.training:
+            original_entropy = entropy(logits)
+            highway_entropy = []
+            highway_logits_all = []
+        if labels is not None:
+            if self.num_labels == 1:
+                #  We are doing regression
+                loss_fct = MSELoss()
+                loss = loss_fct(logits.view(-1), labels.view(-1))
+            else:
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+            # work with highway exits
+            highway_losses = []
+            for highway_exit in outputs[-1]:
+                highway_logits = highway_exit[0]
+                if not self.training:
+                    highway_logits_all.append(highway_logits)
+                    highway_entropy.append(highway_exit[2])
+                if self.num_labels == 1:
+                    #  We are doing regression
+                    loss_fct = MSELoss()
+                    highway_loss = loss_fct(highway_logits.view(-1), labels.view(-1))
+                else:
+                    loss_fct = CrossEntropyLoss()
+                    highway_loss = loss_fct(highway_logits.view(-1, self.num_labels), labels.view(-1))
+                highway_losses.append(highway_loss)
+
+            if train_highway:
+                outputs = (sum(highway_losses[:-1]),) + outputs
+                # exclude the final highway, of course
+            else:
+                outputs = (loss,) + outputs
+        if not self.training:
+            outputs = outputs + ((original_entropy, highway_entropy), exit_layer)
+            if output_layer >= 0:
+                outputs = (
+                    (outputs[0],) + (highway_logits_all[output_layer],) + outputs[2:]
+                )  # use the highway of the last layer
+
+        return outputs  # (loss), logits, (hidden_states), (attentions), (highway_exits)

examples/deebert/src/modeling_highway_roberta.py

@@ -0,0 +1,152 @@
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+import torch.nn as nn
+from torch.nn import CrossEntropyLoss, MSELoss
+
+from transformers.configuration_roberta import RobertaConfig
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.modeling_roberta import ROBERTA_INPUTS_DOCSTRING, ROBERTA_START_DOCSTRING, RobertaEmbeddings
+
+from .modeling_highway_bert import BertPreTrainedModel, DeeBertModel, HighwayException, entropy
+
+
+@add_start_docstrings(
+    "The RoBERTa Model transformer with early exiting (DeeRoBERTa). ",
+    ROBERTA_START_DOCSTRING,
+)
+class DeeRobertaModel(DeeBertModel):
+
+    config_class = RobertaConfig
+    base_model_prefix = "roberta"
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.embeddings = RobertaEmbeddings(config)
+        self.init_weights()
+
+
+@add_start_docstrings(
+    """RoBERTa Model (with early exiting - DeeRoBERTa) with a classifier on top,
+    also takes care of multi-layer training. """,
+    ROBERTA_START_DOCSTRING,
+)
+class DeeRobertaForSequenceClassification(BertPreTrainedModel):
+
+    config_class = RobertaConfig
+    base_model_prefix = "roberta"
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.num_layers = config.num_hidden_layers
+
+        self.roberta = DeeRobertaModel(config)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.classifier = nn.Linear(config.hidden_size, self.config.num_labels)
+
+    @add_start_docstrings_to_callable(ROBERTA_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        labels=None,
+        output_layer=-1,
+        train_highway=False,
+    ):
+        r"""
+            labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`, defaults to :obj:`None`):
+                Labels for computing the sequence classification/regression loss.
+                Indices should be in :obj:`[0, ..., config.num_labels - 1]`.
+                If :obj:`config.num_labels == 1` a regression loss is computed (Mean-Square loss),
+                If :obj:`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+
+        Returns:
+            :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.RobertaConfig`) and inputs:
+            loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`label` is provided):
+                Classification (or regression if config.num_labels==1) loss.
+            logits (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, config.num_labels)`):
+                Classification (or regression if config.num_labels==1) scores (before SoftMax).
+            hidden_states (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``output_hidden_states=True`` is passed or when ``config.output_hidden_states=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer)
+                of shape :obj:`(batch_size, sequence_length, hidden_size)`.
+
+                Hidden-states of the model at the output of each layer plus the initial embedding outputs.
+            attentions (:obj:`tuple(torch.FloatTensor)`, `optional`, returned when ``output_attentions=True`` is passed or when ``config.output_attentions=True``):
+                Tuple of :obj:`torch.FloatTensor` (one for each layer) of shape
+                :obj:`(batch_size, num_heads, sequence_length, sequence_length)`.
+
+                Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
+                heads.
+            highway_exits (:obj:`tuple(tuple(torch.Tensor))`:
+                Tuple of each early exit's results (total length: number of layers)
+                Each tuple is again, a tuple of length 2 - the first entry is logits and the second entry is hidden states.
+        """
+
+        exit_layer = self.num_layers
+        try:
+            outputs = self.roberta(
+                input_ids,
+                attention_mask=attention_mask,
+                token_type_ids=token_type_ids,
+                position_ids=position_ids,
+                head_mask=head_mask,
+                inputs_embeds=inputs_embeds,
+            )
+
+            pooled_output = outputs[1]
+
+            pooled_output = self.dropout(pooled_output)
+            logits = self.classifier(pooled_output)
+            outputs = (logits,) + outputs[2:]  # add hidden states and attention if they are here
+        except HighwayException as e:
+            outputs = e.message
+            exit_layer = e.exit_layer
+            logits = outputs[0]
+
+        if not self.training:
+            original_entropy = entropy(logits)
+            highway_entropy = []
+            highway_logits_all = []
+        if labels is not None:
+            if self.num_labels == 1:
+                #  We are doing regression
+                loss_fct = MSELoss()
+                loss = loss_fct(logits.view(-1), labels.view(-1))
+            else:
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+            # work with highway exits
+            highway_losses = []
+            for highway_exit in outputs[-1]:
+                highway_logits = highway_exit[0]
+                if not self.training:
+                    highway_logits_all.append(highway_logits)
+                    highway_entropy.append(highway_exit[2])
+                if self.num_labels == 1:
+                    #  We are doing regression
+                    loss_fct = MSELoss()
+                    highway_loss = loss_fct(highway_logits.view(-1), labels.view(-1))
+                else:
+                    loss_fct = CrossEntropyLoss()
+                    highway_loss = loss_fct(highway_logits.view(-1, self.num_labels), labels.view(-1))
+                highway_losses.append(highway_loss)
+
+            if train_highway:
+                outputs = (sum(highway_losses[:-1]),) + outputs
+                # exclude the final highway, of course
+            else:
+                outputs = (loss,) + outputs
+        if not self.training:
+            outputs = outputs + ((original_entropy, highway_entropy), exit_layer)
+            if output_layer >= 0:
+                outputs = (
+                    (outputs[0],) + (highway_logits_all[output_layer],) + outputs[2:]
+                )  # use the highway of the last layer
+
+        return outputs  # (loss), logits, (hidden_states), (attentions), entropy

examples/deebert/test_glue_deebert.py

@@ -0,0 +1,98 @@
+import argparse
+import logging
+import sys
+import unittest
+from unittest.mock import patch
+
+import run_glue_deebert
+from transformers.testing_utils import slow
+
+
+logging.basicConfig(level=logging.DEBUG)
+
+logger = logging.getLogger()
+
+
+def get_setup_file():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-f")
+    args = parser.parse_args()
+    return args.f
+
+
+class DeeBertTests(unittest.TestCase):
+    def setup(self) -> None:
+        stream_handler = logging.StreamHandler(sys.stdout)
+        logger.addHandler(stream_handler)
+
+    @slow
+    def test_glue_deebert_train(self):
+
+        train_args = """
+            run_glue_deebert.py
+            --model_type roberta
+            --model_name_or_path roberta-base
+            --task_name MRPC
+            --do_train
+            --do_eval
+            --do_lower_case
+            --data_dir ./tests/fixtures/tests_samples/MRPC/
+            --max_seq_length 128
+            --per_gpu_eval_batch_size=1
+            --per_gpu_train_batch_size=8
+            --learning_rate 2e-4
+            --num_train_epochs 3
+            --overwrite_output_dir
+            --seed 42
+            --output_dir ./examples/deebert/saved_models/roberta-base/MRPC/two_stage
+            --plot_data_dir ./examples/deebert/results/
+            --save_steps 0
+            --overwrite_cache
+            --eval_after_first_stage
+            """.split()
+        with patch.object(sys, "argv", train_args):
+            result = run_glue_deebert.main()
+            for value in result.values():
+                self.assertGreaterEqual(value, 0.666)
+
+        eval_args = """
+            run_glue_deebert.py
+            --model_type roberta
+            --model_name_or_path ./examples/deebert/saved_models/roberta-base/MRPC/two_stage
+            --task_name MRPC
+            --do_eval
+            --do_lower_case
+            --data_dir ./tests/fixtures/tests_samples/MRPC/
+            --output_dir ./examples/deebert/saved_models/roberta-base/MRPC/two_stage
+            --plot_data_dir ./examples/deebert/results/
+            --max_seq_length 128
+            --eval_each_highway
+            --eval_highway
+            --overwrite_cache
+            --per_gpu_eval_batch_size=1
+            """.split()
+        with patch.object(sys, "argv", eval_args):
+            result = run_glue_deebert.main()
+            for value in result.values():
+                self.assertGreaterEqual(value, 0.666)
+
+        entropy_eval_args = """
+            run_glue_deebert.py
+            --model_type roberta
+            --model_name_or_path ./examples/deebert/saved_models/roberta-base/MRPC/two_stage
+            --task_name MRPC
+            --do_eval
+            --do_lower_case
+            --data_dir ./tests/fixtures/tests_samples/MRPC/
+            --output_dir ./examples/deebert/saved_models/roberta-base/MRPC/two_stage
+            --plot_data_dir ./examples/deebert/results/
+            --max_seq_length 128
+            --early_exit_entropy 0.1
+            --eval_highway
+            --overwrite_cache
+            --per_gpu_eval_batch_size=1
+            """.split()
+        with patch.object(sys, "argv", entropy_eval_args):
+            result = run_glue_deebert.main()
+            for value in result.values():
+                self.assertGreaterEqual(value, 0.666)

examples/deebert/train_deebert.sh

@@ -0,0 +1,38 @@
+#!/bin/bash
+export CUDA_VISIBLE_DEVICES=0
+
+PATH_TO_DATA=/h/xinji/projects/GLUE
+
+MODEL_TYPE=bert  # bert or roberta
+MODEL_SIZE=base  # base or large
+DATASET=MRPC  # SST-2, MRPC, RTE, QNLI, QQP, or MNLI
+
+MODEL_NAME=${MODEL_TYPE}-${MODEL_SIZE}
+EPOCHS=10
+if [ $MODEL_TYPE = 'bert' ]
+then
+  EPOCHS=3
+  MODEL_NAME=${MODEL_NAME}-uncased
+fi
+
+
+python -u run_glue_deebert.py \
+  --model_type $MODEL_TYPE \
+  --model_name_or_path $MODEL_NAME \
+  --task_name $DATASET \
+  --do_train \
+  --do_eval \
+  --do_lower_case \
+  --data_dir $PATH_TO_DATA/$DATASET \
+  --max_seq_length 128 \
+  --per_gpu_eval_batch_size=1 \
+  --per_gpu_train_batch_size=8 \
+  --learning_rate 2e-5 \
+  --num_train_epochs $EPOCHS \
+  --overwrite_output_dir \
+  --seed 42 \
+  --output_dir ./saved_models/${MODEL_TYPE}-${MODEL_SIZE}/$DATASET/two_stage \
+  --plot_data_dir ./results/ \
+  --save_steps 0 \
+  --overwrite_cache \
+  --eval_after_first_stage

examples/distillation/README.md

@@ -55,7 +55,7 @@ Here are the results on the *test* sets for 6 of the languages available in XNLI
 
 ## Setup
 
-This part of the library has only be tested with Python3.6+. There are few specific dependencies to install before launching a distillation, you can install them with the command `pip install -r requirements.txt`. 
+This part of the library has only be tested with Python3.6+. There are few specific dependencies to install before launching a distillation, you can install them with the command `pip install -r requirements.txt`.
 
 **Important note:** The training scripts have been updated to support PyTorch v1.2.0 (there are breakings changes compared to v1.1.0).
 
@@ -161,7 +161,7 @@ python -m torch.distributed.launch \
     --master_port $MASTER_PORT \
     train.py \
         --force \
-        --n_gpu $WORLD_SIZE \
+        --gpus $WORLD_SIZE \
         --student_type distilbert \
         --student_config training_configs/distilbert-base-uncased.json \
         --teacher_type bert \

examples/distillation/run_squad_w_distillation.py

@@ -228,14 +228,20 @@ def train(args, train_dataset, model, tokenizer, teacher=None):
                 assert end_logits_tea.size() == end_logits_stu.size()
 
                 loss_fct = nn.KLDivLoss(reduction="batchmean")
-                loss_start = loss_fct(
-                    F.log_softmax(start_logits_stu / args.temperature, dim=-1),
-                    F.softmax(start_logits_tea / args.temperature, dim=-1),
-                ) * (args.temperature ** 2)
-                loss_end = loss_fct(
-                    F.log_softmax(end_logits_stu / args.temperature, dim=-1),
-                    F.softmax(end_logits_tea / args.temperature, dim=-1),
-                ) * (args.temperature ** 2)
+                loss_start = (
+                    loss_fct(
+                        F.log_softmax(start_logits_stu / args.temperature, dim=-1),
+                        F.softmax(start_logits_tea / args.temperature, dim=-1),
+                    )
+                    * (args.temperature ** 2)
+                )
+                loss_end = (
+                    loss_fct(
+                        F.log_softmax(end_logits_stu / args.temperature, dim=-1),
+                        F.softmax(end_logits_tea / args.temperature, dim=-1),
+                    )
+                    * (args.temperature ** 2)
+                )
                 loss_ce = (loss_start + loss_end) / 2.0
 
                 loss = args.alpha_ce * loss_ce + args.alpha_squad * loss

examples/distillation/train.py

@@ -210,7 +210,7 @@ def main():
         help="For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
         "See details at https://nvidia.github.io/apex/amp.html",
     )
-    parser.add_argument("--n_gpu", type=int, default=1, help="Number of GPUs in the node.")
+    parser.add_argument("--gpus", type=int, default=1, help="Number of GPUs in the node.")
     parser.add_argument("--local_rank", type=int, default=-1, help="Distributed training - Local rank")
     parser.add_argument("--seed", type=int, default=56, help="Random seed")
 

examples/distillation/utils.py

@@ -118,7 +118,8 @@ def init_gpu_params(params):
     if params.multi_gpu:
         logger.info("Initializing PyTorch distributed")
         torch.distributed.init_process_group(
-            init_method="env://", backend="nccl",
+            init_method="env://",
+            backend="nccl",
         )