Add BlenderbotTokenizerFast (#13720)

* Add the support for the fast (rust) implementation of BlenbderbotTokenizer

* Fix a converter and a typo in a doc

* Apply the patil-suraj's suggestion

* (Nitpick) Fast tokenization -> Fast Tokenization in doc

* Apply the SaulLu's suggestion

* Apply Narsil's suggestion to fix test pipelines

* Add encoder_no_repeat_ngram_size according to the Narsil's suggestion

* Revert the last (unnecessary) commit

* Override pipeline config for Blenderbot to allow for larger pos. emb.

* make fix-copies
This commit is contained in:
Daniel Stancl
2021-10-29 15:19:01 +02:00
committed by GitHub
parent 5b45422b58
commit d37f1fb8ba
12 changed files with 181 additions and 6 deletions

View File

@@ -379,7 +379,7 @@ Flax), PyTorch, and/or TensorFlow.
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| BigBirdPegasus | ❌ | ❌ | ✅ | ❌ | ❌ |
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| Blenderbot | ✅ | | ✅ | ✅ | ❌ |
| Blenderbot | ✅ | | ✅ | ✅ | ❌ |
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| BlenderbotSmall | ✅ | ✅ | ✅ | ✅ | ❌ |
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+

View File

@@ -81,6 +81,13 @@ BlenderbotTokenizer
:members: build_inputs_with_special_tokens
BlenderbotTokenizerFast
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: transformers.BlenderbotTokenizerFast
:members: build_inputs_with_special_tokens
BlenderbotModel
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~