SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

[Generate] Facilitate PyTorch generate using ModelOutputs (#6735)

Browse Source 
* 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
This commit is contained in:
Patrick von Platen
2020-09-01 12:38:25 +02:00
committed by GitHub
parent 397f819615
commit afc4ece462
20 changed files with 393 additions and 259 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
docs/source/model_doc/encoderdecoder.rst
View File

@@ -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``