Extend Transformers Trainer Class to Enable PyTorch Torchscript for Inference (#17153)

* add jit mode option and model wrap

* Update src/transformers/training_args.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/training_args.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* refine code

* Update src/transformers/trainer.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/trainer.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* add ut and refine code

* code refine

* refine code

* add inference doc

* Update src/transformers/trainer.py

Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

* Update src/transformers/trainer.py

Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

* add cpu inference performance doc

* Update perf_infer_cpu.mdx

* Update perf_infer_cpu.mdx

* Update performance.mdx

* Update _toctree.yml

* refine jit func naming

* Update _toctree.yml

* Delete perf_infer_gpu_one.mdx

* Update perf_infer_cpu.mdx

* Update docs/source/en/perf_infer_cpu.mdx

Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

* add none check before jit

* Update docs/source/en/perf_infer_cpu.mdx

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update docs/source/en/perf_infer_cpu.mdx

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Stas Bekman <stas@stason.org>
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

docs/source/en/_toctree.yml

@@ -87,6 +87,8 @@
     title: Training on many GPUs
   - local: perf_train_cpu
     title: Training on CPU
+  - local: perf_infer_cpu
+    title: Inference on CPU
   - local: perf_hardware
     title: Custom hardware for training
   - local: testing

docs/source/en/perf_infer_cpu.mdx

@@ -0,0 +1,57 @@
+<!--Copyright 2022 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+-->
+
+# Efficient Inference on CPU
+
+This guide focuses on inferencing large models efficiently on CPU.
+
+## PyTorch JIT-mode (TorchScript)
+TorchScript is a way to create serializable and optimizable models from PyTorch code. Any TorchScript program can be saved from a Python process and loaded in a process where there is no Python dependency.
+Comparing to default eager mode, jit mode in PyTorch normally yields better performance for model inference from optimization methodologies like operator fusion.
+
+For a gentle introduction to TorchScript, see the Introduction to [PyTorch TorchScript tutorial](https://pytorch.org/tutorials/beginner/Intro_to_TorchScript_tutorial.html#tracing-modules).
+
+### IPEX Graph Optimization with JIT-mode
+Intel® Extension for PyTorch provides further optimizations in jit mode for Transformers series models. It is highly recommended for users to take advantage of Intel® Extension for PyTorch with jit mode. Some frequently used operator patterns from Transformers models are already supported in Intel® Extension for PyTorch with jit mode fusions. Those fusion patterns like Multi-head-attention fusion, Concat Linear, Linear+Add, Linear+Gelu, Add+LayerNorm fusion and etc. are enabled and perform well. The benefit of the fusion is delivered to users in a transparent fashion. According to the analysis, ~70% of most popular NLP tasks in question-answering, text-classification, and token-classification can get performance benefits with these fusion patterns for both Float32 precision and BFloat16 Mixed precision.
+
+Check more detailed information for [IPEX Graph Optimization](https://intel.github.io/intel-extension-for-pytorch/1.11.200/tutorials/features/graph_optimization.html).
+
+#### IPEX installation:
+
+IPEX release is following PyTorch, check the approaches for [IPEX installation](https://intel.github.io/intel-extension-for-pytorch/).
+
+### Usage of JIT-mode
+To enable jit mode in Trainer, users should add `jit_mode_eval` in Trainer command arguments.
+
+Take an example of the use cases on [Transformers question-answering](https://github.com/huggingface/transformers/tree/main/examples/pytorch/question-answering)
+
+- Inference using jit mode on CPU:
+<pre>python run_qa.py \
+--model_name_or_path csarron/bert-base-uncased-squad-v1 \
+--dataset_name squad \
+--do_eval \
+--max_seq_length 384 \
+--doc_stride 128 \
+--output_dir /tmp/ \
+--no_cuda \
+<b>--jit_mode_eval </b></pre> 
+
+- Inference with IPEX using jit mode on CPU:
+<pre>python run_qa.py \
+--model_name_or_path csarron/bert-base-uncased-squad-v1 \
+--dataset_name squad \
+--do_eval \
+--max_seq_length 384 \
+--doc_stride 128 \
+--output_dir /tmp/ \
+--no_cuda \
+<b>--use_ipex \</b>
+<b>--jit_mode_eval</b></pre> 

docs/source/en/performance.mdx

@@ -58,7 +58,7 @@ Efficient inference with large models in a production environment can be as chal
 
 ### CPU
 
-_Coming soon_
+[Go to CPU inference section](perf_infer_cpu.mdx)
 
 ### Single GPU