dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[BIG] name change

This commit is contained in:
thomwolf 2019-07-05 11:55:36 +02:00
parent 9113b50c96
commit 0bab55d5d5
75 changed files with 280 additions and 230 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
.circleci/config.yml
View File

@ -1,7 +1,7 @@
version: 2
jobs:
build_py3:
working_directory: ~/pytorch-pretrained-BERT
working_directory: ~/pytorch-transformers
docker:
- image: circleci/python:3.5
steps:
@ -10,11 +10,11 @@ jobs:
- run: sudo pip install pytest codecov pytest-cov
- run: sudo pip install spacy ftfy==4.4.3
- run: sudo python -m spacy download en
- run: python -m pytest -sv ./pytorch_pretrained_bert/tests/ --cov
- run: python -m pytest -sv ./pytorch_transformers/tests/ --cov
- run: codecov
parallelism: 4
build_py2:
working_directory: ~/pytorch-pretrained-BERT
working_directory: ~/pytorch-transformers
docker:
- image: circleci/python:2.7
steps:
@ -23,7 +23,7 @@ jobs:
- run: sudo pip install pytest codecov pytest-cov
- run: sudo pip install spacy ftfy==4.4.3
- run: sudo python -m spacy download en
- run: python -m pytest -sv ./pytorch_pretrained_bert/tests/ --cov
- run: python -m pytest -sv ./pytorch_transformers/tests/ --cov
- run: codecov
parallelism: 4
workflows:

2
.coveragerc
View File

@ -1,5 +1,5 @@
[run]
source=pytorch_pretrained_bert
source=pytorch_transformers
[report]
exclude_lines =
pragma: no cover

112
README.md
View File

@ -1,6 +1,6 @@
# PyTorch Pretrained BERT: The Big & Extending Repository of pretrained Transformers
[![CircleCI](https://circleci.com/gh/huggingface/pytorch-pretrained-BERT.svg?style=svg)](https://circleci.com/gh/huggingface/pytorch-pretrained-BERT)
[![CircleCI](https://circleci.com/gh/huggingface/pytorch-transformers.svg?style=svg)](https://circleci.com/gh/huggingface/pytorch-transformers)
This repository contains op-for-op PyTorch reimplementations, pre-trained models and fine-tuning examples for:
@ -47,7 +47,7 @@ This repo was tested on Python 2.7 and 3.5+ (examples are tested only on python
PyTorch pretrained bert can be installed by pip as follows:
```bash
pip install pytorch-pretrained-bert
pip install pytorch-transformers
```
If you want to reproduce the original tokenization process of the `OpenAI GPT` paper, you will need to install `ftfy` (limit to version 4.4.3 if you are using Python 2) and `SpaCy` :
@ -73,7 +73,7 @@ python -m spacy download en
Again, if you don't install `ftfy` and `SpaCy`, the `OpenAI GPT` tokenizer will default to tokenize using BERT's `BasicTokenizer` followed by Byte-Pair Encoding (which should be fine for most usage).
A series of tests is included in the [tests folder](https://github.com/huggingface/pytorch-pretrained-BERT/tree/master/tests) and can be run using `pytest` (install pytest if needed: `pip install pytest`).
A series of tests is included in the [tests folder](https://github.com/huggingface/pytorch-transformers/tree/master/tests) and can be run using `pytest` (install pytest if needed: `pip install pytest`).
You can run the tests with the command:
```bash
@ -84,51 +84,51 @@ python -m pytest -sv tests/
This package comprises the following classes that can be imported in Python and are detailed in the [Doc](#doc) section of this readme:
- Eight **Bert** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling.py`](./pytorch_pretrained_bert/modeling.py) file):
- [`BertModel`](./pytorch_pretrained_bert/modeling.py#L639) - raw BERT Transformer model (**fully pre-trained**),
- [`BertForMaskedLM`](./pytorch_pretrained_bert/modeling.py#L793) - BERT Transformer with the pre-trained masked language modeling head on top (**fully pre-trained**),
- [`BertForNextSentencePrediction`](./pytorch_pretrained_bert/modeling.py#L854) - BERT Transformer with the pre-trained next sentence prediction classifier on top (**fully pre-trained**),
- [`BertForPreTraining`](./pytorch_pretrained_bert/modeling.py#L722) - BERT Transformer with masked language modeling head and next sentence prediction classifier on top (**fully pre-trained**),
- [`BertForSequenceClassification`](./pytorch_pretrained_bert/modeling.py#L916) - BERT Transformer with a sequence classification head on top (BERT Transformer is **pre-trained**, the sequence classification head **is only initialized and has to be trained**),
- [`BertForMultipleChoice`](./pytorch_pretrained_bert/modeling.py#L982) - BERT Transformer with a multiple choice head on top (used for task like Swag) (BERT Transformer is **pre-trained**, the multiple choice classification head **is only initialized and has to be trained**),
- [`BertForTokenClassification`](./pytorch_pretrained_bert/modeling.py#L1051) - BERT Transformer with a token classification head on top (BERT Transformer is **pre-trained**, the token classification head **is only initialized and has to be trained**),
- [`BertForQuestionAnswering`](./pytorch_pretrained_bert/modeling.py#L1124) - BERT Transformer with a token classification head on top (BERT Transformer is **pre-trained**, the token classification head **is only initialized and has to be trained**).
- Eight **Bert** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling.py`](./pytorch_transformers/modeling.py) file):
- [`BertModel`](./pytorch_transformers/modeling.py#L639) - raw BERT Transformer model (**fully pre-trained**),
- [`BertForMaskedLM`](./pytorch_transformers/modeling.py#L793) - BERT Transformer with the pre-trained masked language modeling head on top (**fully pre-trained**),
- [`BertForNextSentencePrediction`](./pytorch_transformers/modeling.py#L854) - BERT Transformer with the pre-trained next sentence prediction classifier on top (**fully pre-trained**),
- [`BertForPreTraining`](./pytorch_transformers/modeling.py#L722) - BERT Transformer with masked language modeling head and next sentence prediction classifier on top (**fully pre-trained**),
- [`BertForSequenceClassification`](./pytorch_transformers/modeling.py#L916) - BERT Transformer with a sequence classification head on top (BERT Transformer is **pre-trained**, the sequence classification head **is only initialized and has to be trained**),
- [`BertForMultipleChoice`](./pytorch_transformers/modeling.py#L982) - BERT Transformer with a multiple choice head on top (used for task like Swag) (BERT Transformer is **pre-trained**, the multiple choice classification head **is only initialized and has to be trained**),
- [`BertForTokenClassification`](./pytorch_transformers/modeling.py#L1051) - BERT Transformer with a token classification head on top (BERT Transformer is **pre-trained**, the token classification head **is only initialized and has to be trained**),
- [`BertForQuestionAnswering`](./pytorch_transformers/modeling.py#L1124) - BERT Transformer with a token classification head on top (BERT Transformer is **pre-trained**, the token classification head **is only initialized and has to be trained**).
- Three **OpenAI GPT** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling_openai.py`](./pytorch_pretrained_bert/modeling_openai.py) file):
- [`OpenAIGPTModel`](./pytorch_pretrained_bert/modeling_openai.py#L536) - raw OpenAI GPT Transformer model (**fully pre-trained**),
- [`OpenAIGPTLMHeadModel`](./pytorch_pretrained_bert/modeling_openai.py#L643) - OpenAI GPT Transformer with the tied language modeling head on top (**fully pre-trained**),
- [`OpenAIGPTDoubleHeadsModel`](./pytorch_pretrained_bert/modeling_openai.py#L722) - OpenAI GPT Transformer with the tied language modeling head and a multiple choice classification head on top (OpenAI GPT Transformer is **pre-trained**, the multiple choice classification head **is only initialized and has to be trained**),
- Three **OpenAI GPT** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling_openai.py`](./pytorch_transformers/modeling_openai.py) file):
- [`OpenAIGPTModel`](./pytorch_transformers/modeling_openai.py#L536) - raw OpenAI GPT Transformer model (**fully pre-trained**),
- [`OpenAIGPTLMHeadModel`](./pytorch_transformers/modeling_openai.py#L643) - OpenAI GPT Transformer with the tied language modeling head on top (**fully pre-trained**),
- [`OpenAIGPTDoubleHeadsModel`](./pytorch_transformers/modeling_openai.py#L722) - OpenAI GPT Transformer with the tied language modeling head and a multiple choice classification head on top (OpenAI GPT Transformer is **pre-trained**, the multiple choice classification head **is only initialized and has to be trained**),
- Two **Transformer-XL** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling_transfo_xl.py`](./pytorch_pretrained_bert/modeling_transfo_xl.py) file):
- [`TransfoXLModel`](./pytorch_pretrained_bert/modeling_transfo_xl.py#L983) - Transformer-XL model which outputs the last hidden state and memory cells (**fully pre-trained**),
- [`TransfoXLLMHeadModel`](./pytorch_pretrained_bert/modeling_transfo_xl.py#L1260) - Transformer-XL with the tied adaptive softmax head on top for language modeling which outputs the logits/loss and memory cells (**fully pre-trained**),
- Two **Transformer-XL** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling_transfo_xl.py`](./pytorch_transformers/modeling_transfo_xl.py) file):
- [`TransfoXLModel`](./pytorch_transformers/modeling_transfo_xl.py#L983) - Transformer-XL model which outputs the last hidden state and memory cells (**fully pre-trained**),
- [`TransfoXLLMHeadModel`](./pytorch_transformers/modeling_transfo_xl.py#L1260) - Transformer-XL with the tied adaptive softmax head on top for language modeling which outputs the logits/loss and memory cells (**fully pre-trained**),
- Three **OpenAI GPT-2** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling_gpt2.py`](./pytorch_pretrained_bert/modeling_gpt2.py) file):
- [`GPT2Model`](./pytorch_pretrained_bert/modeling_gpt2.py#L479) - raw OpenAI GPT-2 Transformer model (**fully pre-trained**),
- [`GPT2LMHeadModel`](./pytorch_pretrained_bert/modeling_gpt2.py#L559) - OpenAI GPT-2 Transformer with the tied language modeling head on top (**fully pre-trained**),
- [`GPT2DoubleHeadsModel`](./pytorch_pretrained_bert/modeling_gpt2.py#L624) - OpenAI GPT-2 Transformer with the tied language modeling head and a multiple choice classification head on top (OpenAI GPT-2 Transformer is **pre-trained**, the multiple choice classification head **is only initialized and has to be trained**),
- Three **OpenAI GPT-2** PyTorch models (`torch.nn.Module`) with pre-trained weights (in the [`modeling_gpt2.py`](./pytorch_transformers/modeling_gpt2.py) file):
- [`GPT2Model`](./pytorch_transformers/modeling_gpt2.py#L479) - raw OpenAI GPT-2 Transformer model (**fully pre-trained**),
- [`GPT2LMHeadModel`](./pytorch_transformers/modeling_gpt2.py#L559) - OpenAI GPT-2 Transformer with the tied language modeling head on top (**fully pre-trained**),
- [`GPT2DoubleHeadsModel`](./pytorch_transformers/modeling_gpt2.py#L624) - OpenAI GPT-2 Transformer with the tied language modeling head and a multiple choice classification head on top (OpenAI GPT-2 Transformer is **pre-trained**, the multiple choice classification head **is only initialized and has to be trained**),
- Tokenizers for **BERT** (using word-piece) (in the [`tokenization.py`](./pytorch_pretrained_bert/tokenization.py) file):
- Tokenizers for **BERT** (using word-piece) (in the [`tokenization.py`](./pytorch_transformers/tokenization.py) file):
- `BasicTokenizer` - basic tokenization (punctuation splitting, lower casing, etc.),
- `WordpieceTokenizer` - WordPiece tokenization,
- `BertTokenizer` - perform end-to-end tokenization, i.e. basic tokenization followed by WordPiece tokenization.
- Tokenizer for **OpenAI GPT** (using Byte-Pair-Encoding) (in the [`tokenization_openai.py`](./pytorch_pretrained_bert/tokenization_openai.py) file):
- Tokenizer for **OpenAI GPT** (using Byte-Pair-Encoding) (in the [`tokenization_openai.py`](./pytorch_transformers/tokenization_openai.py) file):
- `OpenAIGPTTokenizer` - perform Byte-Pair-Encoding (BPE) tokenization.
- Tokenizer for **Transformer-XL** (word tokens ordered by frequency for adaptive softmax) (in the [`tokenization_transfo_xl.py`](./pytorch_pretrained_bert/tokenization_transfo_xl.py) file):
- Tokenizer for **Transformer-XL** (word tokens ordered by frequency for adaptive softmax) (in the [`tokenization_transfo_xl.py`](./pytorch_transformers/tokenization_transfo_xl.py) file):
- `OpenAIGPTTokenizer` - perform word tokenization and can order words by frequency in a corpus for use in an adaptive softmax.
- Tokenizer for **OpenAI GPT-2** (using byte-level Byte-Pair-Encoding) (in the [`tokenization_gpt2.py`](./pytorch_pretrained_bert/tokenization_gpt2.py) file):
- Tokenizer for **OpenAI GPT-2** (using byte-level Byte-Pair-Encoding) (in the [`tokenization_gpt2.py`](./pytorch_transformers/tokenization_gpt2.py) file):
- `GPT2Tokenizer` - perform byte-level Byte-Pair-Encoding (BPE) tokenization.
- Optimizer for **BERT** (in the [`optimization.py`](./pytorch_pretrained_bert/optimization.py) file):
- Optimizer for **BERT** (in the [`optimization.py`](./pytorch_transformers/optimization.py) file):
- `BertAdam` - Bert version of Adam algorithm with weight decay fix, warmup and linear decay of the learning rate.
- Optimizer for **OpenAI GPT** (in the [`optimization_openai.py`](./pytorch_pretrained_bert/optimization_openai.py) file):
- Optimizer for **OpenAI GPT** (in the [`optimization_openai.py`](./pytorch_transformers/optimization_openai.py) file):
- `OpenAIAdam` - OpenAI GPT version of Adam algorithm with weight decay fix, warmup and linear decay of the learning rate.
- Configuration classes for BERT, OpenAI GPT and Transformer-XL (in the respective [`modeling.py`](./pytorch_pretrained_bert/modeling.py), [`modeling_openai.py`](./pytorch_pretrained_bert/modeling_openai.py), [`modeling_transfo_xl.py`](./pytorch_pretrained_bert/modeling_transfo_xl.py) files):
- Configuration classes for BERT, OpenAI GPT and Transformer-XL (in the respective [`modeling.py`](./pytorch_transformers/modeling.py), [`modeling_openai.py`](./pytorch_transformers/modeling_openai.py), [`modeling_transfo_xl.py`](./pytorch_transformers/modeling_transfo_xl.py) files):
- `BertConfig` - Configuration class to store the configuration of a `BertModel` with utilities to read and write from JSON configuration files.
- `OpenAIGPTConfig` - Configuration class to store the configuration of a `OpenAIGPTModel` with utilities to read and write from JSON configuration files.
- `GPT2Config` - Configuration class to store the configuration of a `GPT2Model` with utilities to read and write from JSON configuration files.
@ -175,7 +175,7 @@ First let's prepare a tokenized input with `BertTokenizer`
```python
import torch
from pytorch_pretrained_bert import BertTokenizer, BertModel, BertForMaskedLM
from pytorch_transformers import BertTokenizer, BertModel, BertForMaskedLM
# OPTIONAL: if you want to have more information on what's happening, activate the logger as follows
import logging
@ -252,7 +252,7 @@ First let's prepare a tokenized input with `OpenAIGPTTokenizer`
```python
import torch
from pytorch_pretrained_bert import OpenAIGPTTokenizer, OpenAIGPTModel, OpenAIGPTLMHeadModel
from pytorch_transformers import OpenAIGPTTokenizer, OpenAIGPTModel, OpenAIGPTLMHeadModel
# OPTIONAL: if you want to have more information on what's happening, activate the logger as follows
import logging
@ -339,7 +339,7 @@ First let's prepare a tokenized input with `TransfoXLTokenizer`
```python
import torch
from pytorch_pretrained_bert import TransfoXLTokenizer, TransfoXLModel, TransfoXLLMHeadModel
from pytorch_transformers import TransfoXLTokenizer, TransfoXLModel, TransfoXLLMHeadModel
# OPTIONAL: if you want to have more information on what's happening, activate the logger as follows
import logging
@ -414,7 +414,7 @@ First let's prepare a tokenized input with `GPT2Tokenizer`
```python
import torch
from pytorch_pretrained_bert import GPT2Tokenizer, GPT2Model, GPT2LMHeadModel
from pytorch_transformers import GPT2Tokenizer, GPT2Model, GPT2LMHeadModel
# OPTIONAL: if you want to have more information on what's happening, activate the logger as follows
import logging
@ -552,7 +552,7 @@ where
- `bert_config.json` or `openai_gpt_config.json` a configuration file for the model, and
- `pytorch_model.bin` a PyTorch dump of a pre-trained instance of `BertForPreTraining`, `OpenAIGPTModel`, `TransfoXLModel`, `GPT2LMHeadModel` (saved with the usual `torch.save()`)
If `PRE_TRAINED_MODEL_NAME_OR_PATH` is a shortcut name, the pre-trained weights will be downloaded from AWS S3 (see the links [here](pytorch_pretrained_bert/modeling.py)) and stored in a cache folder to avoid future download (the cache folder can be found at `~/.pytorch_pretrained_bert/`).
If `PRE_TRAINED_MODEL_NAME_OR_PATH` is a shortcut name, the pre-trained weights will be downloaded from AWS S3 (see the links [here](pytorch_transformers/modeling.py)) and stored in a cache folder to avoid future download (the cache folder can be found at `~/.pytorch_transformers/`).
- `cache_dir` can be an optional path to a specific directory to download and cache the pre-trained model weights. This option is useful in particular when you are using distributed training: to avoid concurrent access to the same weights you can set for example `cache_dir='./pretrained_model_{}'.format(args.local_rank)` (see the section on distributed training for more information).
- `from_tf`: should we load the weights from a locally saved TensorFlow checkpoint
@ -586,19 +586,19 @@ model = GPT2Model.from_pretrained('gpt2')
#### Cache directory
`pytorch_pretrained_bert` save the pretrained weights in a cache directory which is located at (in this order of priority):
`pytorch_transformers` save the pretrained weights in a cache directory which is located at (in this order of priority):
- `cache_dir` optional arguments to the `from_pretrained()` method (see above),
- shell environment variable `PYTORCH_PRETRAINED_BERT_CACHE`,
- PyTorch cache home + `/pytorch_pretrained_bert/`
- PyTorch cache home + `/pytorch_transformers/`
where PyTorch cache home is defined by (in this order):
- shell environment variable `ENV_TORCH_HOME`
- shell environment variable `ENV_XDG_CACHE_HOME` + `/torch/`)
- default: `~/.cache/torch/`
Usually, if you don't set any specific environment variable, `pytorch_pretrained_bert` cache will be at `~/.cache/torch/pytorch_pretrained_bert/`.
Usually, if you don't set any specific environment variable, `pytorch_transformers` cache will be at `~/.cache/torch/pytorch_transformers/`.
You can alsways safely delete `pytorch_pretrained_bert` cache but the pretrained model weights and vocabulary files wil have to be re-downloaded from our S3.
You can alsways safely delete `pytorch_transformers` cache but the pretrained model weights and vocabulary files wil have to be re-downloaded from our S3.
### Serialization best-practices
@ -621,7 +621,7 @@ The *default filenames* of these files are as follow:
Here is the recommended way of saving the model, configuration and vocabulary to an `output_dir` directory and reloading the model and tokenizer afterwards:
```python
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
output_dir = "./models/"
@ -719,7 +719,7 @@ The model can be instantiated with the following arguments:
The inputs and output are **identical to the TensorFlow model inputs and outputs**.
We detail them here. This model takes as *inputs*:
[`modeling.py`](./pytorch_pretrained_bert/modeling.py)
[`modeling.py`](./pytorch_transformers/modeling.py)
- `input_ids`: a torch.LongTensor of shape [batch_size, sequence_length] with the word token indices in the vocabulary (see the tokens preprocessing logic in the scripts [`run_bert_extract_features.py`](./examples/run_bert_extract_features.py), [`run_bert_classifier.py`](./examples/run_bert_classifier.py) and [`run_bert_squad.py`](./examples/run_bert_squad.py)), and
- `token_type_ids`: an optional torch.LongTensor of shape [batch_size, sequence_length] with the token types indices selected in [0, 1]. Type 0 corresponds to a `sentence A` and type 1 corresponds to a `sentence B` token (see BERT paper for more details).
- `attention_mask`: an optional torch.LongTensor of shape [batch_size, sequence_length] with indices selected in [0, 1]. It's a mask to be used if some input sequence lengths are smaller than the max input sequence length of the current batch. It's the mask that we typically use for attention when a batch has varying length sentences.
@ -852,7 +852,7 @@ The model can be instantiated with the following arguments:
The inputs and output are **identical to the TensorFlow model inputs and outputs**.
We detail them here. This model takes as *inputs*:
[`modeling_openai.py`](./pytorch_pretrained_bert/modeling_openai.py)
[`modeling_openai.py`](./pytorch_transformers/modeling_openai.py)
- `input_ids`: a torch.LongTensor of shape [batch_size, sequence_length] (or more generally [d_1, ..., d_n, sequence_length] were d_1 ... d_n are arbitrary dimensions) with the word BPE token indices selected in the range [0, total_tokens_embeddings[
- `position_ids`: an optional torch.LongTensor with the same shape as input_ids
with the position indices (selected in the range [0, config.n_positions - 1[.
@ -905,7 +905,7 @@ Transformer XL use a relative positioning with sinusiodal patterns and adaptive
- the tokens in the vocabulary have to be sorted to decreasing frequency.
This model takes as *inputs*:
[`modeling_transfo_xl.py`](./pytorch_pretrained_bert/modeling_transfo_xl.py)
[`modeling_transfo_xl.py`](./pytorch_transformers/modeling_transfo_xl.py)
- `input_ids`: a torch.LongTensor of shape [batch_size, sequence_length] with the token indices selected in the range [0, self.config.n_token[
- `mems`: an optional memory of hidden states from previous forward passes as a list (num layers) of hidden states at the entry of each layer. Each hidden states has shape [self.config.mem_len, bsz, self.config.d_model]. Note that the first two dimensions are transposed in `mems` with regards to `input_ids`.
@ -952,7 +952,7 @@ The model can be instantiated with the following arguments:
The inputs and output are **identical to the TensorFlow model inputs and outputs**.
We detail them here. This model takes as *inputs*:
[`modeling_gpt2.py`](./pytorch_pretrained_bert/modeling_gpt2.py)
[`modeling_gpt2.py`](./pytorch_transformers/modeling_gpt2.py)
- `input_ids`: a torch.LongTensor of shape [batch_size, sequence_length] (or more generally [d_1, ..., d_n, sequence_length] were d_1 ... d_n are arbitrary dimensions) with the word BPE token indices selected in the range [0, vocab_size[
- `position_ids`: an optional torch.LongTensor with the same shape as input_ids
with the position indices (selected in the range [0, config.n_positions - 1[.
@ -1020,7 +1020,7 @@ and three methods:
- `convert_ids_to_tokens(tokens)`: convert a list of `int` indices in a list of `str` tokens in the vocabulary.
- `save_vocabulary(directory_path)`: save the vocabulary file to `directory_path`. Return the path to the saved vocabulary file: `vocab_file_path`. The vocabulary can be reloaded with `BertTokenizer.from_pretrained('vocab_file_path')` or `BertTokenizer.from_pretrained('directory_path')`.
Please refer to the doc strings and code in [`tokenization.py`](./pytorch_pretrained_bert/tokenization.py) for the details of the `BasicTokenizer` and `WordpieceTokenizer` classes. In general it is recommended to use `BertTokenizer` unless you know what you are doing.
Please refer to the doc strings and code in [`tokenization.py`](./pytorch_transformers/tokenization.py) for the details of the `BasicTokenizer` and `WordpieceTokenizer` classes. In general it is recommended to use `BertTokenizer` unless you know what you are doing.
#### `OpenAIGPTTokenizer`
@ -1043,7 +1043,7 @@ and five methods:
- `decode(ids, skip_special_tokens=False, clean_up_tokenization_spaces=False)`: decode a list of `int` indices in a string and do some post-processing if needed: (i) remove special tokens from the output and (ii) clean up tokenization spaces.
- `save_vocabulary(directory_path)`: save the vocabulary, merge and special tokens files to `directory_path`. Return the path to the three files: `vocab_file_path`, `merge_file_path`, `special_tokens_file_path`. The vocabulary can be reloaded with `OpenAIGPTTokenizer.from_pretrained('directory_path')`.
Please refer to the doc strings and code in [`tokenization_openai.py`](./pytorch_pretrained_bert/tokenization_openai.py) for the details of the `OpenAIGPTTokenizer`.
Please refer to the doc strings and code in [`tokenization_openai.py`](./pytorch_transformers/tokenization_openai.py) for the details of the `OpenAIGPTTokenizer`.
#### `TransfoXLTokenizer`
@ -1051,7 +1051,7 @@ Please refer to the doc strings and code in [`tokenization_openai.py`](./pytorch
The API is similar to the API of `BertTokenizer` (see above).
Please refer to the doc strings and code in [`tokenization_transfo_xl.py`](./pytorch_pretrained_bert/tokenization_transfo_xl.py) for the details of these additional methods in `TransfoXLTokenizer`.
Please refer to the doc strings and code in [`tokenization_transfo_xl.py`](./pytorch_transformers/tokenization_transfo_xl.py) for the details of these additional methods in `TransfoXLTokenizer`.
#### `GPT2Tokenizer`
@ -1073,7 +1073,7 @@ and two methods:
- `decode(tokens)`: convert back a list of `int` tokens in a `str`.
- `save_vocabulary(directory_path)`: save the vocabulary, merge and special tokens files to `directory_path`. Return the path to the three files: `vocab_file_path`, `merge_file_path`, `special_tokens_file_path`. The vocabulary can be reloaded with `OpenAIGPTTokenizer.from_pretrained('directory_path')`.
Please refer to [`tokenization_gpt2.py`](./pytorch_pretrained_bert/tokenization_gpt2.py) for more details on the `GPT2Tokenizer`.
Please refer to [`tokenization_gpt2.py`](./pytorch_transformers/tokenization_gpt2.py) for more details on the `GPT2Tokenizer`.
### Optimizers
@ -1155,7 +1155,7 @@ Here is how to use these techniques in our scripts:
- **Distributed training**: Distributed training can be activated by supplying an integer greater or equal to 0 to the `--local_rank` argument (see below).
- **16-bits training**: 16-bits training, also called mixed-precision training, can reduce the memory requirement of your model on the GPU by using half-precision training, basically allowing to double the batch size. If you have a recent GPU (starting from NVIDIA Volta architecture) you should see no decrease in speed. A good introduction to Mixed precision training can be found [here](https://devblogs.nvidia.com/mixed-precision-training-deep-neural-networks/) and a full documentation is [here](https://docs.nvidia.com/deeplearning/sdk/mixed-precision-training/index.html). In our scripts, this option can be activated by setting the `--fp16` flag and you can play with loss scaling using the `--loss_scale` flag (see the previously linked documentation for details on loss scaling). The loss scale can be zero in which case the scale is dynamically adjusted or a positive power of two in which case the scaling is static.
To use 16-bits training and distributed training, you need to install NVIDIA's apex extension [as detailed here](https://github.com/nvidia/apex). You will find more information regarding the internals of `apex` and how to use `apex` in [the doc and the associated repository](https://github.com/nvidia/apex). The results of the tests performed on pytorch-BERT by the NVIDIA team (and my trials at reproducing them) can be consulted in [the relevant PR of the present repository](https://github.com/huggingface/pytorch-pretrained-BERT/pull/116).
To use 16-bits training and distributed training, you need to install NVIDIA's apex extension [as detailed here](https://github.com/nvidia/apex). You will find more information regarding the internals of `apex` and how to use `apex` in [the doc and the associated repository](https://github.com/nvidia/apex). The results of the tests performed on pytorch-BERT by the NVIDIA team (and my trials at reproducing them) can be consulted in [the relevant PR of the present repository](https://github.com/huggingface/pytorch-transformers/pull/116).
Note: To use *Distributed Training*, you will need to run one training script on each of your machines. This can be done for example by running the following command on each server (see [the above mentioned blog post]((https://medium.com/huggingface/training-larger-batches-practical-tips-on-1-gpu-multi-gpu-distributed-setups-ec88c3e51255)) for more details):
```bash
@ -1660,7 +1660,7 @@ To help you understand and use these features, we have added a specific example
## Notebooks
We include [three Jupyter Notebooks](https://github.com/huggingface/pytorch-pretrained-BERT/tree/master/notebooks) that can be used to check that the predictions of the PyTorch model are identical to the predictions of the original TensorFlow model.
We include [three Jupyter Notebooks](https://github.com/huggingface/pytorch-transformers/tree/master/notebooks) that can be used to check that the predictions of the PyTorch model are identical to the predictions of the original TensorFlow model.
- The first NoteBook ([Comparing-TF-and-PT-models.ipynb](./notebooks/Comparing-TF-and-PT-models.ipynb)) extracts the hidden states of a full sequence on each layers of the TensorFlow and the PyTorch models and computes the standard deviation between them. In the given example, we get a standard deviation of 1.5e-7 to 9e-7 on the various hidden state of the models.
@ -1676,7 +1676,7 @@ A command-line interface is provided to convert a TensorFlow checkpoint in a PyT
### BERT
You can convert any TensorFlow checkpoint for BERT (in particular [the pre-trained models released by Google](https://github.com/google-research/bert#pre-trained-models)) in a PyTorch save file by using the [`convert_tf_checkpoint_to_pytorch.py`](./pytorch_pretrained_bert/convert_tf_checkpoint_to_pytorch.py ) script.
You can convert any TensorFlow checkpoint for BERT (in particular [the pre-trained models released by Google](https://github.com/google-research/bert#pre-trained-models)) in a PyTorch save file by using the [`convert_tf_checkpoint_to_pytorch.py`](./pytorch_transformers/convert_tf_checkpoint_to_pytorch.py ) script.
This CLI takes as input a TensorFlow checkpoint (three files starting with `bert_model.ckpt`) and the associated configuration file (`bert_config.json`), and creates a PyTorch model for this configuration, loads the weights from the TensorFlow checkpoint in the PyTorch model and saves the resulting model in a standard PyTorch save file that can be imported using `torch.load()` (see examples in [`run_bert_extract_features.py`](./examples/run_bert_extract_features.py), [`run_bert_classifier.py`](./examples/run_bert_classifier.py) and [`run_bert_squad.py`](./examples/run_bert_squad.py)).
@ -1689,7 +1689,7 @@ Here is an example of the conversion process for a pre-trained `BERT-Base Uncase
```shell
export BERT_BASE_DIR=/path/to/bert/uncased_L-12_H-768_A-12
pytorch_pretrained_bert bert \
pytorch_transformers bert \
$BERT_BASE_DIR/bert_model.ckpt \
$BERT_BASE_DIR/bert_config.json \
$BERT_BASE_DIR/pytorch_model.bin
@ -1704,7 +1704,7 @@ Here is an example of the conversion process for a pre-trained OpenAI GPT model,
```shell
export OPENAI_GPT_CHECKPOINT_FOLDER_PATH=/path/to/openai/pretrained/numpy/weights
pytorch_pretrained_bert gpt \
pytorch_transformers gpt \
$OPENAI_GPT_CHECKPOINT_FOLDER_PATH \
$PYTORCH_DUMP_OUTPUT \
[OPENAI_GPT_CONFIG]
@ -1717,7 +1717,7 @@ Here is an example of the conversion process for a pre-trained Transformer-XL mo
```shell
export TRANSFO_XL_CHECKPOINT_FOLDER_PATH=/path/to/transfo/xl/checkpoint
pytorch_pretrained_bert transfo_xl \
pytorch_transformers transfo_xl \
$TRANSFO_XL_CHECKPOINT_FOLDER_PATH \
$PYTORCH_DUMP_OUTPUT \
[TRANSFO_XL_CONFIG]
@ -1730,7 +1730,7 @@ Here is an example of the conversion process for a pre-trained OpenAI's GPT-2 mo
```shell
export GPT2_DIR=/path/to/gpt2/checkpoint
pytorch_pretrained_bert gpt2 \
pytorch_transformers gpt2 \
$GPT2_DIR/model.ckpt \
$PYTORCH_DUMP_OUTPUT \
[GPT2_CONFIG]
@ -1744,7 +1744,7 @@ Here is an example of the conversion process for a pre-trained XLNet model, fine
export TRANSFO_XL_CHECKPOINT_PATH=/path/to/xlnet/checkpoint
export TRANSFO_XL_CONFIG_PATH=/path/to/xlnet/config
pytorch_pretrained_bert xlnet \
pytorch_transformers xlnet \
$TRANSFO_XL_CHECKPOINT_PATH \
$TRANSFO_XL_CONFIG_PATH \
$PYTORCH_DUMP_OUTPUT \

2
docker/Dockerfile
View File

@ -2,6 +2,6 @@ FROM pytorch/pytorch:latest
RUN git clone https://github.com/NVIDIA/apex.git && cd apex && python setup.py install --cuda_ext --cpp_ext
RUN pip install pytorch-pretrained-bert
RUN pip install pytorch_transformers
WORKDIR /workspace

2
examples/bertology.py
View File

@ -12,7 +12,7 @@ from torch.utils.data import DataLoader, SequentialSampler, TensorDataset, Subse
from torch.utils.data.distributed import DistributedSampler
from torch.nn import CrossEntropyLoss, MSELoss
from pytorch_pretrained_bert import BertForSequenceClassification, BertTokenizer
from pytorch_transformers import BertForSequenceClassification, BertTokenizer
from utils_glue import processors, output_modes, convert_examples_to_features, compute_metrics

2
examples/generation_xlnet.py
View File

@ -1,6 +1,6 @@
import torch
from torch.nn import functional as F
from pytorch_pretrained_bert import XLNetModel, XLNetLMHeadModel, XLNetTokenizer
from pytorch_transformers import XLNetModel, XLNetLMHeadModel, XLNetTokenizer
import logging
logging.basicConfig(level=logging.INFO)

8
examples/lm_finetuning/finetune_on_pregenerated.py
View File

@ -13,10 +13,10 @@ from torch.utils.data import DataLoader, Dataset, RandomSampler
from torch.utils.data.distributed import DistributedSampler
from tqdm import tqdm
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_bert import BertForPreTraining
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_bert import BertForPreTraining
from pytorch_transformers.tokenization_bert import BertTokenizer
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
InputFeatures = namedtuple("InputFeatures", "input_ids input_mask segment_ids lm_label_ids is_next")

2
examples/lm_finetuning/pregenerate_training_data.py
View File

@ -5,7 +5,7 @@ from tempfile import TemporaryDirectory
import shelve
from random import random, randrange, randint, shuffle, choice
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_transformers.tokenization_bert import BertTokenizer
import numpy as np
import json
import collections

8
examples/lm_finetuning/simple_lm_finetuning.py
View File

@ -29,10 +29,10 @@ from torch.utils.data import DataLoader, Dataset, RandomSampler
from torch.utils.data.distributed import DistributedSampler
from tqdm import tqdm, trange
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_bert import BertForPreTraining
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_bert import BertForPreTraining
from pytorch_transformers.tokenization_bert import BertTokenizer
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
logging.basicConfig(format='%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt='%m/%d/%Y %H:%M:%S',

8
examples/run_bert_classifier.py
View File

@ -34,10 +34,10 @@ from torch.nn import CrossEntropyLoss, MSELoss
from tensorboardX import SummaryWriter
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_bert import BertForSequenceClassification
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_bert import BertForSequenceClassification
from pytorch_transformers.tokenization_bert import BertTokenizer
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
from utils_glue import processors, output_modes, convert_examples_to_features, compute_metrics

4
examples/run_bert_extract_features.py
View File

@ -28,8 +28,8 @@ import torch
from torch.utils.data import TensorDataset, DataLoader, SequentialSampler
from torch.utils.data.distributed import DistributedSampler
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_pretrained_bert.modeling_bert import BertModel
from pytorch_transformers.tokenization_bert import BertTokenizer
from pytorch_transformers.modeling_bert import BertModel
logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt = '%m/%d/%Y %H:%M:%S',

8
examples/run_bert_squad.py
View File

@ -33,10 +33,10 @@ from tqdm import tqdm, trange
from tensorboardX import SummaryWriter
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_bert import BertForQuestionAnswering
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_bert import BertForQuestionAnswering
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers.tokenization_bert import BertTokenizer
from utils_squad import read_squad_examples, convert_examples_to_features, RawResult, write_predictions

8
examples/run_bert_swag.py
View File

@ -32,10 +32,10 @@ from torch.utils.data import (DataLoader, RandomSampler, SequentialSampler,
from torch.utils.data.distributed import DistributedSampler
from tqdm import tqdm, trange
from pytorch_pretrained_bert.file_utils import PYTORCH_PRETRAINED_BERT_CACHE, WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_bert import BertForMultipleChoice, BertConfig
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_transformers.file_utils import PYTORCH_PRETRAINED_BERT_CACHE, WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_bert import BertForMultipleChoice, BertConfig
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers.tokenization_bert import BertTokenizer
logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt = '%m/%d/%Y %H:%M:%S',

2
examples/run_gpt2.py
View File

@ -8,7 +8,7 @@ import torch
import torch.nn.functional as F
import numpy as np
from pytorch_pretrained_bert import GPT2LMHeadModel, GPT2Tokenizer
from pytorch_transformers import GPT2LMHeadModel, GPT2Tokenizer
logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt = '%m/%d/%Y %H:%M:%S',

2
examples/run_openai_gpt.py
View File

@ -39,7 +39,7 @@ import torch
from torch.utils.data import (DataLoader, RandomSampler, SequentialSampler,
TensorDataset)
from pytorch_pretrained_bert import (OpenAIGPTDoubleHeadsModel, OpenAIGPTTokenizer,
from pytorch_transformers import (OpenAIGPTDoubleHeadsModel, OpenAIGPTTokenizer,
OpenAIAdam, cached_path, WEIGHTS_NAME, CONFIG_NAME)
ROCSTORIES_URL = "https://s3.amazonaws.com/datasets.huggingface.co/ROCStories.tar.gz"

2
examples/run_transfo_xl.py
View File

@ -28,7 +28,7 @@ import math
import torch
from pytorch_pretrained_bert import TransfoXLLMHeadModel, TransfoXLCorpus, TransfoXLTokenizer
from pytorch_transformers import TransfoXLLMHeadModel, TransfoXLCorpus, TransfoXLTokenizer
logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt = '%m/%d/%Y %H:%M:%S',

8
examples/run_xlnet_classifier.py
View File

@ -34,10 +34,10 @@ from torch.nn import CrossEntropyLoss, MSELoss
from tensorboardX import SummaryWriter
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_xlnet import XLNetForSequenceClassification
from pytorch_pretrained_bert.tokenization_xlnet import XLNetTokenizer
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_xlnet import XLNetForSequenceClassification
from pytorch_transformers.tokenization_xlnet import XLNetTokenizer
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
from utils_glue import processors, output_modes, convert_examples_to_features, compute_metrics

8
examples/run_xlnet_squad.py
View File

@ -33,10 +33,10 @@ from tqdm import tqdm, trange
from tensorboardX import SummaryWriter
from pytorch_pretrained_bert import WEIGHTS_NAME, CONFIG_NAME
from pytorch_pretrained_bert.modeling_xlnet import BertForQuestionAnswering
from pytorch_pretrained_bert.tokenization_xlnet import XLNetTokenizer
from pytorch_pretrained_bert.optimization import BertAdam, WarmupLinearSchedule
from pytorch_transformers import WEIGHTS_NAME, CONFIG_NAME
from pytorch_transformers.modeling_xlnet import BertForQuestionAnswering
from pytorch_transformers.tokenization_xlnet import XLNetTokenizer
from pytorch_transformers.optimization import BertAdam, WarmupLinearSchedule
from utils_squad import read_squad_examples, convert_examples_to_features, RawResult, write_predictions

50
examples/tests/examples_tests.py Normal file
View File

@ -0,0 +1,50 @@
# coding=utf-8
# Copyright 2018 HuggingFace Inc..
#
# 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 specific language governing permissions and
# limitations under the License.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
import unittest
import json
import random
import shutil
import pytest
import torch
from pytorch_transformers import PretrainedConfig, PreTrainedModel
from pytorch_transformers.modeling_bert import BertModel, BertConfig, PRETRAINED_MODEL_ARCHIVE_MAP, PRETRAINED_CONFIG_ARCHIVE_MAP
class ModelUtilsTest(unittest.TestCase):
def test_model_from_pretrained(self):
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
config = BertConfig.from_pretrained(model_name)
self.assertIsNotNone(config)
self.assertIsInstance(config, PretrainedConfig)
model = BertModel.from_pretrained(model_name)
self.assertIsNotNone(model)
self.assertIsInstance(model, PreTrainedModel)
config = BertConfig.from_pretrained(model_name, output_attentions=True, output_hidden_states=True)
model = BertModel.from_pretrained(model_name, output_attentions=True, output_hidden_states=True)
self.assertEqual(model.config.output_attentions, True)
self.assertEqual(model.config.output_hidden_states, True)
self.assertEqual(model.config, config)
if __name__ == "__main__":
unittest.main()

2
examples/utils_squad.py
View File

@ -24,7 +24,7 @@ import math
import collections
from io import open
from pytorch_pretrained_bert.tokenization_bert import BasicTokenizer, whitespace_tokenize
from pytorch_transformers.tokenization_bert import BasicTokenizer, whitespace_tokenize
logger = logging.getLogger(__name__)

38
hubconfs/bert_hubconf.py
View File

@ -1,5 +1,5 @@
from pytorch_pretrained_bert.tokenization_bert import BertTokenizer
from pytorch_pretrained_bert.modeling_bert import (
from pytorch_transformers.tokenization_bert import BertTokenizer
from pytorch_transformers.modeling_bert import (
BertModel,
BertForNextSentencePrediction,
BertForMaskedLM,
@ -86,7 +86,7 @@ def bertTokenizer(*args, **kwargs):
Example:
>>> import torch
>>> sentence = 'Hello, World!'
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> toks = tokenizer.tokenize(sentence)
['Hello', '##,', 'World', '##!']
>>> ids = tokenizer.convert_tokens_to_ids(toks)
@ -106,7 +106,7 @@ def bertModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -115,7 +115,7 @@ def bertModel(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertModel', 'bert-base-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertModel', 'bert-base-cased')
>>> model.eval()
# Predict hidden states features for each layer
>>> with torch.no_grad():
@ -135,7 +135,7 @@ def bertForNextSentencePrediction(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -144,7 +144,7 @@ def bertForNextSentencePrediction(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertForNextSentencePrediction
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForNextSentencePrediction', 'bert-base-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForNextSentencePrediction', 'bert-base-cased')
>>> model.eval()
# Predict the next sentence classification logits
>>> with torch.no_grad():
@ -165,7 +165,7 @@ def bertForPreTraining(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -173,7 +173,7 @@ def bertForPreTraining(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertForPreTraining
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForPreTraining', 'bert-base-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForPreTraining', 'bert-base-cased')
>>> masked_lm_logits_scores, seq_relationship_logits = model(tokens_tensor, segments_tensors)
"""
model = BertForPreTraining.from_pretrained(*args, **kwargs)
@ -189,7 +189,7 @@ def bertForMaskedLM(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -200,7 +200,7 @@ def bertForMaskedLM(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertForMaskedLM
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForMaskedLM', 'bert-base-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForMaskedLM', 'bert-base-cased')
>>> model.eval()
# Predict all tokens
>>> with torch.no_grad():
@ -231,7 +231,7 @@ def bertForSequenceClassification(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -240,7 +240,7 @@ def bertForSequenceClassification(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertForSequenceClassification
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForSequenceClassification', 'bert-base-cased', num_labels=2)
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForSequenceClassification', 'bert-base-cased', num_labels=2)
>>> model.eval()
# Predict the sequence classification logits
>>> with torch.no_grad():
@ -266,7 +266,7 @@ def bertForMultipleChoice(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -275,7 +275,7 @@ def bertForMultipleChoice(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens, indexed_tokens]).unsqueeze(0)
>>> segments_tensors = torch.tensor([segments_ids, segments_ids]).unsqueeze(0)
# Load bertForMultipleChoice
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForMultipleChoice', 'bert-base-cased', num_choices=2)
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForMultipleChoice', 'bert-base-cased', num_choices=2)
>>> model.eval()
# Predict the multiple choice logits
>>> with torch.no_grad():
@ -299,7 +299,7 @@ def bertForQuestionAnswering(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -308,7 +308,7 @@ def bertForQuestionAnswering(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertForQuestionAnswering
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForQuestionAnswering', 'bert-base-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForQuestionAnswering', 'bert-base-cased')
>>> model.eval()
# Predict the start and end positions logits
>>> with torch.no_grad():
@ -338,7 +338,7 @@ def bertForTokenClassification(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'bertTokenizer', 'bert-base-cased', do_basic_tokenize=False)
# Prepare tokenized input
>>> text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
>>> tokenized_text = tokenizer.tokenize(text)
@ -347,7 +347,7 @@ def bertForTokenClassification(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
>>> segments_tensors = torch.tensor([segments_ids])
# Load bertForTokenClassification
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'bertForTokenClassification', 'bert-base-cased', num_labels=2)
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'bertForTokenClassification', 'bert-base-cased', num_labels=2)
>>> model.eval()
# Predict the token classification logits
>>> with torch.no_grad():

18
hubconfs/gpt2_hubconf.py
View File

@ -1,5 +1,5 @@
from pytorch_pretrained_bert.tokenization_gpt2 import GPT2Tokenizer
from pytorch_pretrained_bert.modeling_gpt2 import (
from pytorch_transformers.tokenization_gpt2 import GPT2Tokenizer
from pytorch_transformers.modeling_gpt2 import (
GPT2Model,
GPT2LMHeadModel,
GPT2DoubleHeadsModel
@ -53,7 +53,7 @@ def gpt2Tokenizer(*args, **kwargs):
Example:
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2Tokenizer', 'gpt2')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'gpt2Tokenizer', 'gpt2')
>>> text = "Who was Jim Henson ?"
>>> indexed_tokens = tokenizer.encode(tokenized_text)
@ -72,7 +72,7 @@ def gpt2Model(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2Tokenizer', 'gpt2')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'gpt2Tokenizer', 'gpt2')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -83,7 +83,7 @@ def gpt2Model(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load gpt2Model
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2Model', 'gpt2')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'gpt2Model', 'gpt2')
>>> model.eval()
# Predict hidden states features for each layer
@ -105,7 +105,7 @@ def gpt2LMHeadModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2Tokenizer', 'gpt2')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'gpt2Tokenizer', 'gpt2')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -116,7 +116,7 @@ def gpt2LMHeadModel(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load gpt2LMHeadModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2LMHeadModel', 'gpt2')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'gpt2LMHeadModel', 'gpt2')
>>> model.eval()
# Predict hidden states features for each layer
@ -144,7 +144,7 @@ def gpt2DoubleHeadsModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2Tokenizer', 'gpt2')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'gpt2Tokenizer', 'gpt2')
# Prepare tokenized input
>>> text1 = "Who was Jim Henson ? Jim Henson was a puppeteer"
@ -157,7 +157,7 @@ def gpt2DoubleHeadsModel(*args, **kwargs):
>>> mc_token_ids = torch.LongTensor([[len(tokenized_text1)-1, len(tokenized_text2)-1]])
# Load gpt2DoubleHeadsModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'gpt2DoubleHeadsModel', 'gpt2')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'gpt2DoubleHeadsModel', 'gpt2')
>>> model.eval()
# Predict hidden states features for each layer

18
hubconfs/gpt_hubconf.py
View File

@ -1,5 +1,5 @@
from pytorch_pretrained_bert.tokenization_openai import OpenAIGPTTokenizer
from pytorch_pretrained_bert.modeling_openai import (
from pytorch_transformers.tokenization_openai import OpenAIGPTTokenizer
from pytorch_transformers.modeling_openai import (
OpenAIGPTModel,
OpenAIGPTLMHeadModel,
OpenAIGPTDoubleHeadsModel
@ -77,7 +77,7 @@ def openAIGPTTokenizer(*args, **kwargs):
Example:
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTTokenizer', 'openai-gpt')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTTokenizer', 'openai-gpt')
>>> text = "Who was Jim Henson ? Jim Henson was a puppeteer"
>>> tokenized_text = tokenizer.tokenize(text)
@ -98,7 +98,7 @@ def openAIGPTModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTTokenizer', 'openai-gpt')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTTokenizer', 'openai-gpt')
# Prepare tokenized input
>>> text = "Who was Jim Henson ? Jim Henson was a puppeteer"
@ -107,7 +107,7 @@ def openAIGPTModel(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
# Load openAIGPTModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTModel', 'openai-gpt')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTModel', 'openai-gpt')
>>> model.eval()
# Predict hidden states features for each layer
@ -127,7 +127,7 @@ def openAIGPTLMHeadModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTTokenizer', 'openai-gpt')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTTokenizer', 'openai-gpt')
# Prepare tokenized input
>>> text = "Who was Jim Henson ? Jim Henson was a puppeteer"
@ -136,7 +136,7 @@ def openAIGPTLMHeadModel(*args, **kwargs):
>>> tokens_tensor = torch.tensor([indexed_tokens])
# Load openAIGPTLMHeadModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTLMHeadModel', 'openai-gpt')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTLMHeadModel', 'openai-gpt')
>>> model.eval()
# Predict hidden states features for each layer
@ -162,7 +162,7 @@ def openAIGPTDoubleHeadsModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTTokenizer', 'openai-gpt')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTTokenizer', 'openai-gpt')
# Prepare tokenized input
>>> text1 = "Who was Jim Henson ? Jim Henson was a puppeteer"
@ -175,7 +175,7 @@ def openAIGPTDoubleHeadsModel(*args, **kwargs):
>>> mc_token_ids = torch.LongTensor([[len(tokenized_text1)-1, len(tokenized_text2)-1]])
# Load openAIGPTDoubleHeadsModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'openAIGPTDoubleHeadsModel', 'openai-gpt')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'openAIGPTDoubleHeadsModel', 'openai-gpt')
>>> model.eval()
# Predict hidden states features for each layer

14
hubconfs/transformer_xl_hubconf.py
View File

@ -1,5 +1,5 @@
from pytorch_pretrained_bert.tokenization_transfo_xl import TransfoXLTokenizer
from pytorch_pretrained_bert.modeling_transfo_xl import (
from pytorch_transformers.tokenization_transfo_xl import TransfoXLTokenizer
from pytorch_transformers.modeling_transfo_xl import (
TransfoXLModel,
TransfoXLLMHeadModel
)
@ -46,7 +46,7 @@ def transformerXLTokenizer(*args, **kwargs):
Example:
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'transformerXLTokenizer', 'transfo-xl-wt103')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'transformerXLTokenizer', 'transfo-xl-wt103')
>>> text = "Who was Jim Henson ?"
>>> tokenized_text = tokenizer.tokenize(tokenized_text)
@ -64,7 +64,7 @@ def transformerXLModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'transformerXLTokenizer', 'transfo-xl-wt103')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'transformerXLTokenizer', 'transfo-xl-wt103')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -77,7 +77,7 @@ def transformerXLModel(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load transformerXLModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'transformerXLModel', 'transfo-xl-wt103')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'transformerXLModel', 'transfo-xl-wt103')
>>> model.eval()
# Predict hidden states features for each layer
@ -99,7 +99,7 @@ def transformerXLLMHeadModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'transformerXLTokenizer', 'transfo-xl-wt103')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'transformerXLTokenizer', 'transfo-xl-wt103')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -112,7 +112,7 @@ def transformerXLLMHeadModel(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load transformerXLLMHeadModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'transformerXLLMHeadModel', 'transfo-xl-wt103')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'transformerXLLMHeadModel', 'transfo-xl-wt103')
>>> model.eval()
# Predict hidden states features for each layer

16
hubconfs/xlm_hubconf.py
View File

@ -1,5 +1,5 @@
from pytorch_pretrained_bert.tokenization_xlm import XLMTokenizer
from pytorch_pretrained_bert.modeling_xlm import (
from pytorch_transformers.tokenization_xlm import XLMTokenizer
from pytorch_transformers.modeling_xlm import (
XLMConfig,
XLMModel,
XLMWithLMHeadModel,
@ -18,7 +18,7 @@ xlm_start_docstring = """
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlmTokenizer', 'xlm-mlm-en-2048')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlmTokenizer', 'xlm-mlm-en-2048')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -77,7 +77,7 @@ def xlmTokenizer(*args, **kwargs):
Example:
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlmTokenizer', 'xlm-mlm-en-2048')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlmTokenizer', 'xlm-mlm-en-2048')
>>> text = "Who was Jim Henson ?"
>>> indexed_tokens = tokenizer.encode(tokenized_text)
@ -91,7 +91,7 @@ def xlmTokenizer(*args, **kwargs):
def xlmModel(*args, **kwargs):
"""
# Load xlmModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlmModel', 'xlm-mlm-en-2048')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'xlmModel', 'xlm-mlm-en-2048')
>>> model.eval()
# Predict hidden states features for each layer
@ -116,7 +116,7 @@ def xlmLMHeadModel(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load xlnetLMHeadModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetLMHeadModel', 'xlm-mlm-en-2048')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'xlnetLMHeadModel', 'xlm-mlm-en-2048')
>>> model.eval()
# Predict hidden states features for each layer
@ -143,7 +143,7 @@ def xlmLMHeadModel(*args, **kwargs):
# Example:
# # Load the tokenizer
# >>> import torch
# >>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetTokenizer', 'xlm-mlm-en-2048')
# >>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlnetTokenizer', 'xlm-mlm-en-2048')
# # Prepare tokenized input
# >>> text1 = "Who was Jim Henson ? Jim Henson was a puppeteer"
@ -156,7 +156,7 @@ def xlmLMHeadModel(*args, **kwargs):
# >>> mc_token_ids = torch.LongTensor([[len(tokenized_text1)-1, len(tokenized_text2)-1]])
# # Load xlnetForSequenceClassification
# >>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetForSequenceClassification', 'xlm-mlm-en-2048')
# >>> model = torch.hub.load('huggingface/pytorch-transformers', 'xlnetForSequenceClassification', 'xlm-mlm-en-2048')
# >>> model.eval()
# # Predict sequence classes logits

18
hubconfs/xlnet_hubconf.1.py
View File

@ -1,5 +1,5 @@
from pytorch_pretrained_bert.tokenization_xlnet import XLNetTokenizer
from pytorch_pretrained_bert.modeling_xlnet import (
from pytorch_transformers.tokenization_xlnet import XLNetTokenizer
from pytorch_transformers.modeling_xlnet import (
XLNetConfig,
XLNetModel,
XLNetLMHeadModel,
@ -54,7 +54,7 @@ def xlnetTokenizer(*args, **kwargs):
Example:
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetTokenizer', 'xlnet-large-cased')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlnetTokenizer', 'xlnet-large-cased')
>>> text = "Who was Jim Henson ?"
>>> indexed_tokens = tokenizer.encode(tokenized_text)
@ -73,7 +73,7 @@ def xlnetModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetTokenizer', 'xlnet-large-cased')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlnetTokenizer', 'xlnet-large-cased')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -84,7 +84,7 @@ def xlnetModel(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load xlnetModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetModel', 'xlnet-large-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'xlnetModel', 'xlnet-large-cased')
>>> model.eval()
# Predict hidden states features for each layer
@ -107,7 +107,7 @@ def xlnetLMHeadModel(*args, **kwargs):
Example:
# Load the tokenizer
>>> import torch
>>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetTokenizer', 'xlnet-large-cased')
>>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlnetTokenizer', 'xlnet-large-cased')
# Prepare tokenized input
>>> text_1 = "Who was Jim Henson ?"
@ -118,7 +118,7 @@ def xlnetLMHeadModel(*args, **kwargs):
>>> tokens_tensor_2 = torch.tensor([indexed_tokens_2])
# Load xlnetLMHeadModel
>>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetLMHeadModel', 'xlnet-large-cased')
>>> model = torch.hub.load('huggingface/pytorch-transformers', 'xlnetLMHeadModel', 'xlnet-large-cased')
>>> model.eval()
# Predict hidden states features for each layer
@ -145,7 +145,7 @@ def xlnetLMHeadModel(*args, **kwargs):
# Example:
# # Load the tokenizer
# >>> import torch
# >>> tokenizer = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetTokenizer', 'xlnet-large-cased')
# >>> tokenizer = torch.hub.load('huggingface/pytorch-transformers', 'xlnetTokenizer', 'xlnet-large-cased')
# # Prepare tokenized input
# >>> text1 = "Who was Jim Henson ? Jim Henson was a puppeteer"
@ -158,7 +158,7 @@ def xlnetLMHeadModel(*args, **kwargs):
# >>> mc_token_ids = torch.LongTensor([[len(tokenized_text1)-1, len(tokenized_text2)-1]])
# # Load xlnetForSequenceClassification
# >>> model = torch.hub.load('huggingface/pytorch-pretrained-BERT', 'xlnetForSequenceClassification', 'xlnet-large-cased')
# >>> model = torch.hub.load('huggingface/pytorch-transformers', 'xlnetForSequenceClassification', 'xlnet-large-cased')
# >>> model.eval()
# # Predict sequence classes logits

8
notebooks/Comparing-TF-and-PT-models-MLM-NSP.ipynb
View File

@ -78,7 +78,7 @@
"import importlib.util\n",
"import sys\n",
"import tensorflow as tf\n",
"import pytorch_pretrained_bert as ppb\n",
"import pytorch_transformers as ppb\n",
"\n",
"def del_all_flags(FLAGS):\n",
" flags_dict = FLAGS._flags() \n",
@ -3997,9 +3997,9 @@
"name": "stderr",
"output_type": "stream",
"text": [
"11/16/2018 11:03:05 - INFO - pytorch_pretrained_bert.modeling_bert - loading archive file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased.tar.gz from cache at /Users/thomaswolf/.pytorch_pretrained_bert/9c41111e2de84547a463fd39217199738d1e3deb72d4fec4399e6e241983c6f0.ae3cef932725ca7a30cdcb93fc6e09150a55e2a130ec7af63975a16c153ae2ba\n",
"11/16/2018 11:03:05 - INFO - pytorch_pretrained_bert.modeling_bert - extracting archive file /Users/thomaswolf/.pytorch_pretrained_bert/9c41111e2de84547a463fd39217199738d1e3deb72d4fec4399e6e241983c6f0.ae3cef932725ca7a30cdcb93fc6e09150a55e2a130ec7af63975a16c153ae2ba to temp dir /var/folders/yx/cw8n_njx3js5jksyw_qlp8p00000gn/T/tmpaqgsm566\n",
"11/16/2018 11:03:08 - INFO - pytorch_pretrained_bert.modeling_bert - Model config {\n",
"11/16/2018 11:03:05 - INFO - pytorch_transformers.modeling_bert - loading archive file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased.tar.gz from cache at /Users/thomaswolf/.pytorch_transformers/9c41111e2de84547a463fd39217199738d1e3deb72d4fec4399e6e241983c6f0.ae3cef932725ca7a30cdcb93fc6e09150a55e2a130ec7af63975a16c153ae2ba\n",
"11/16/2018 11:03:05 - INFO - pytorch_transformers.modeling_bert - extracting archive file /Users/thomaswolf/.pytorch_transformers/9c41111e2de84547a463fd39217199738d1e3deb72d4fec4399e6e241983c6f0.ae3cef932725ca7a30cdcb93fc6e09150a55e2a130ec7af63975a16c153ae2ba to temp dir /var/folders/yx/cw8n_njx3js5jksyw_qlp8p00000gn/T/tmpaqgsm566\n",
"11/16/2018 11:03:08 - INFO - pytorch_transformers.modeling_bert - Model config {\n",
" \"attention_probs_dropout_prob\": 0.1,\n",
" \"hidden_act\": \"gelu\",\n",
" \"hidden_dropout_prob\": 0.1,\n",

6
notebooks/Comparing-TF-and-PT-models.ipynb
View File

@ -342,7 +342,7 @@
"outputs": [],
"source": [
"import extract_features\n",
"import pytorch_pretrained_bert as ppb\n",
"import pytorch_transformers as ppb\n",
"from extract_features import *"
]
},
@ -375,8 +375,8 @@
"name": "stderr",
"output_type": "stream",
"text": [
"11/15/2018 16:21:18 - INFO - pytorch_pretrained_bert.modeling_bert - loading archive file ../../google_models/uncased_L-12_H-768_A-12/\n",
"11/15/2018 16:21:18 - INFO - pytorch_pretrained_bert.modeling_bert - Model config {\n",
"11/15/2018 16:21:18 - INFO - pytorch_transformers.modeling_bert - loading archive file ../../google_models/uncased_L-12_H-768_A-12/\n",
"11/15/2018 16:21:18 - INFO - pytorch_transformers.modeling_bert - Model config {\n",
" \"attention_probs_dropout_prob\": 0.1,\n",
" \"hidden_act\": \"gelu\",\n",
" \"hidden_dropout_prob\": 0.1,\n",

2
pytorch_pretrained_bert/__init__.py → pytorch_transformers/__init__.py
View File

@ -1,4 +1,4 @@
__version__ = "0.6.2"
__version__ = "0.7.0"
from .tokenization_bert import BertTokenizer, BasicTokenizer, WordpieceTokenizer
from .tokenization_openai import OpenAIGPTTokenizer
from .tokenization_transfo_xl import (TransfoXLTokenizer, TransfoXLCorpus)

28
pytorch_pretrained_bert/__main__.py → pytorch_transformers/__main__.py
View File

@ -4,24 +4,24 @@ def main():
if (len(sys.argv) < 4 or len(sys.argv) > 6) or sys.argv[1] not in ["bert", "gpt", "transfo_xl", "gpt2", "xlnet"]:
print(
"Should be used as one of: \n"
">> `pytorch_pretrained_bert bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`, \n"
">> `pytorch_pretrained_bert gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG]`, \n"
">> `pytorch_pretrained_bert transfo_xl TF_CHECKPOINT_OR_DATASET PYTORCH_DUMP_OUTPUT [TF_CONFIG]` or \n"
">> `pytorch_pretrained_bert gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [GPT2_CONFIG]` or \n"
">> `pytorch_pretrained_bert xlnet TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT [FINETUNING_TASK_NAME]`")
">> `pytorch_transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`, \n"
">> `pytorch_transformers gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG]`, \n"
">> `pytorch_transformers transfo_xl TF_CHECKPOINT_OR_DATASET PYTORCH_DUMP_OUTPUT [TF_CONFIG]` or \n"
">> `pytorch_transformers gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [GPT2_CONFIG]` or \n"
">> `pytorch_transformers xlnet TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT [FINETUNING_TASK_NAME]`")
else:
if sys.argv[1] == "bert":
try:
from .convert_tf_checkpoint_to_pytorch import convert_tf_checkpoint_to_pytorch
except ImportError:
print("pytorch_pretrained_bert can only be used from the commandline to convert TensorFlow models in PyTorch, "
print("pytorch_transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
"In that case, it requires TensorFlow to be installed. Please see "
"https://www.tensorflow.org/install/ for installation instructions.")
raise
if len(sys.argv) != 5:
# pylint: disable=line-too-long
print("Should be used as `pytorch_pretrained_bert bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`")
print("Should be used as `pytorch_transformers bert TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT`")
else:
PYTORCH_DUMP_OUTPUT = sys.argv.pop()
TF_CONFIG = sys.argv.pop()
@ -31,7 +31,7 @@ def main():
from .convert_openai_checkpoint_to_pytorch import convert_openai_checkpoint_to_pytorch
if len(sys.argv) < 4 or len(sys.argv) > 5:
# pylint: disable=line-too-long
print("Should be used as `pytorch_pretrained_bert gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG]`")
print("Should be used as `pytorch_transformers gpt OPENAI_GPT_CHECKPOINT_FOLDER_PATH PYTORCH_DUMP_OUTPUT [OPENAI_GPT_CONFIG]`")
else:
OPENAI_GPT_CHECKPOINT_FOLDER_PATH = sys.argv[2]
PYTORCH_DUMP_OUTPUT = sys.argv[3]
@ -46,13 +46,13 @@ def main():
try:
from .convert_transfo_xl_checkpoint_to_pytorch import convert_transfo_xl_checkpoint_to_pytorch
except ImportError:
print("pytorch_pretrained_bert can only be used from the commandline to convert TensorFlow models in PyTorch, "
print("pytorch_transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
"In that case, it requires TensorFlow to be installed. Please see "
"https://www.tensorflow.org/install/ for installation instructions.")
raise
if len(sys.argv) < 4 or len(sys.argv) > 5:
# pylint: disable=line-too-long
print("Should be used as `pytorch_pretrained_bert transfo_xl TF_CHECKPOINT/TF_DATASET_FILE PYTORCH_DUMP_OUTPUT [TF_CONFIG]`")
print("Should be used as `pytorch_transformers transfo_xl TF_CHECKPOINT/TF_DATASET_FILE PYTORCH_DUMP_OUTPUT [TF_CONFIG]`")
else:
if 'ckpt' in sys.argv[2].lower():
TF_CHECKPOINT = sys.argv[2]
@ -70,14 +70,14 @@ def main():
try:
from .convert_gpt2_checkpoint_to_pytorch import convert_gpt2_checkpoint_to_pytorch
except ImportError:
print("pytorch_pretrained_bert can only be used from the commandline to convert TensorFlow models in PyTorch, "
print("pytorch_transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
"In that case, it requires TensorFlow to be installed. Please see "
"https://www.tensorflow.org/install/ for installation instructions.")
raise
if len(sys.argv) < 4 or len(sys.argv) > 5:
# pylint: disable=line-too-long
print("Should be used as `pytorch_pretrained_bert gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [TF_CONFIG]`")
print("Should be used as `pytorch_transformers gpt2 TF_CHECKPOINT PYTORCH_DUMP_OUTPUT [TF_CONFIG]`")
else:
TF_CHECKPOINT = sys.argv[2]
PYTORCH_DUMP_OUTPUT = sys.argv[3]
@ -90,14 +90,14 @@ def main():
try:
from .convert_xlnet_checkpoint_to_pytorch import convert_xlnet_checkpoint_to_pytorch
except ImportError:
print("pytorch_pretrained_bert can only be used from the commandline to convert TensorFlow models in PyTorch, "
print("pytorch_transformers can only be used from the commandline to convert TensorFlow models in PyTorch, "
"In that case, it requires TensorFlow to be installed. Please see "
"https://www.tensorflow.org/install/ for installation instructions.")
raise
if len(sys.argv) < 5 or len(sys.argv) > 6:
# pylint: disable=line-too-long
print("Should be used as `pytorch_pretrained_bert xlnet TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT [FINETUNING_TASK_NAME]`")
print("Should be used as `pytorch_transformers xlnet TF_CHECKPOINT TF_CONFIG PYTORCH_DUMP_OUTPUT [FINETUNING_TASK_NAME]`")
else:
TF_CHECKPOINT = sys.argv[2]
TF_CONFIG = sys.argv[3]

2
pytorch_pretrained_bert/convert_gpt2_checkpoint_to_pytorch.py → pytorch_transformers/convert_gpt2_checkpoint_to_pytorch.py
View File

@ -21,7 +21,7 @@ from io import open
import torch
from pytorch_pretrained_bert.modeling_gpt2 import (CONFIG_NAME, WEIGHTS_NAME,
from pytorch_transformers.modeling_gpt2 import (CONFIG_NAME, WEIGHTS_NAME,
GPT2Config,
GPT2Model,
load_tf_weights_in_gpt2)

2
pytorch_pretrained_bert/convert_openai_checkpoint_to_pytorch.py → pytorch_transformers/convert_openai_checkpoint_to_pytorch.py
View File

@ -21,7 +21,7 @@ from io import open
import torch
from pytorch_pretrained_bert.modeling_openai import (CONFIG_NAME, WEIGHTS_NAME,
from pytorch_transformers.modeling_openai import (CONFIG_NAME, WEIGHTS_NAME,
OpenAIGPTConfig,
OpenAIGPTModel,
load_tf_weights_in_openai_gpt)

2
pytorch_pretrained_bert/convert_tf_checkpoint_to_pytorch.py → pytorch_transformers/convert_tf_checkpoint_to_pytorch.py
View File

@ -25,7 +25,7 @@ import tensorflow as tf
import torch
import numpy as np
from pytorch_pretrained_bert.modeling_bert import BertConfig, BertForPreTraining, load_tf_weights_in_bert
from pytorch_transformers.modeling_bert import BertConfig, BertForPreTraining, load_tf_weights_in_bert
def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_file, pytorch_dump_path):
# Initialise PyTorch model

6
pytorch_pretrained_bert/convert_transfo_xl_checkpoint_to_pytorch.py → pytorch_transformers/convert_transfo_xl_checkpoint_to_pytorch.py
View File

@ -23,13 +23,13 @@ from io import open
import torch
import pytorch_pretrained_bert.tokenization_transfo_xl as data_utils
from pytorch_pretrained_bert.modeling_transfo_xl import (CONFIG_NAME,
import pytorch_transformers.tokenization_transfo_xl as data_utils
from pytorch_transformers.modeling_transfo_xl import (CONFIG_NAME,
WEIGHTS_NAME,
TransfoXLConfig,
TransfoXLLMHeadModel,
load_tf_weights_in_transfo_xl)
from pytorch_pretrained_bert.tokenization_transfo_xl import (CORPUS_NAME,
from pytorch_transformers.tokenization_transfo_xl import (CORPUS_NAME,
VOCAB_NAME)
if sys.version_info[0] == 2:

4
pytorch_pretrained_bert/convert_xlm_checkpoint_to_pytorch.py → pytorch_transformers/convert_xlm_checkpoint_to_pytorch.py
View File

@ -23,8 +23,8 @@ from io import open
import torch
import numpy
from pytorch_pretrained_bert.modeling_xlm import (CONFIG_NAME, WEIGHTS_NAME, XLMConfig, XLMModel)
from pytorch_pretrained_bert.tokenization_xlm import MERGES_NAME, VOCAB_NAME
from pytorch_transformers.modeling_xlm import (CONFIG_NAME, WEIGHTS_NAME, XLMConfig, XLMModel)
from pytorch_transformers.tokenization_xlm import MERGES_NAME, VOCAB_NAME
def convert_xlm_checkpoint_to_pytorch(xlm_checkpoint_path, pytorch_dump_folder_path):

2
pytorch_pretrained_bert/convert_xlnet_checkpoint_to_pytorch.py → pytorch_transformers/convert_xlnet_checkpoint_to_pytorch.py
View File

@ -22,7 +22,7 @@ import os
import argparse
import torch
from pytorch_pretrained_bert.modeling_xlnet import (CONFIG_NAME, WEIGHTS_NAME,
from pytorch_transformers.modeling_xlnet import (CONFIG_NAME, WEIGHTS_NAME,
XLNetConfig,
XLNetLMHeadModel, XLNetForQuestionAnswering,
XLNetForSequenceClassification,

2
pytorch_pretrained_bert/file_utils.py → pytorch_transformers/file_utils.py
View File

@ -29,7 +29,7 @@ except ImportError:
torch_cache_home = os.path.expanduser(
os.getenv('TORCH_HOME', os.path.join(
os.getenv('XDG_CACHE_HOME', '~/.cache'), 'torch')))
default_cache_path = os.path.join(torch_cache_home, 'pytorch_pretrained_bert')
default_cache_path = os.path.join(torch_cache_home, 'pytorch_transformers')
try:
from urllib.parse import urlparse

0
pytorch_pretrained_bert/model_utils.py → pytorch_transformers/model_utils.py
View File

0
pytorch_pretrained_bert/modeling_bert.py → pytorch_transformers/modeling_bert.py
View File

0
pytorch_pretrained_bert/modeling_gpt2.py → pytorch_transformers/modeling_gpt2.py
View File

0
pytorch_pretrained_bert/modeling_openai.py → pytorch_transformers/modeling_openai.py
View File

0
pytorch_pretrained_bert/modeling_transfo_xl.py → pytorch_transformers/modeling_transfo_xl.py
View File

0
pytorch_pretrained_bert/modeling_transfo_xl_utilities.py → pytorch_transformers/modeling_transfo_xl_utilities.py
View File

2
pytorch_pretrained_bert/modeling_xlm.py → pytorch_transformers/modeling_xlm.py
View File

@ -204,7 +204,7 @@ def gelu(x):
GELU activation
https://arxiv.org/abs/1606.08415
https://github.com/huggingface/pytorch-openai-transformer-lm/blob/master/model_pytorch.py#L14
https://github.com/huggingface/pytorch-pretrained-BERT/blob/master/modeling.py
https://github.com/huggingface/pytorch-transformers/blob/master/modeling.py
"""
# return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3))))
return 0.5 * x * (1.0 + torch.erf(x / math.sqrt(2.0)))

0
pytorch_pretrained_bert/modeling_xlnet.py → pytorch_transformers/modeling_xlnet.py
View File

0
pytorch_pretrained_bert/optimization.py → pytorch_transformers/optimization.py
View File

0
pytorch_pretrained_bert/optimization_openai.py → pytorch_transformers/optimization_openai.py
View File

0
pytorch_pretrained_bert/tests/__init__.py → pytorch_transformers/tests/__init__.py
View File

0
pytorch_pretrained_bert/tests/conftest.py → pytorch_transformers/tests/conftest.py
View File

0
pytorch_pretrained_bert/tests/fixtures/input.txt → pytorch_transformers/tests/fixtures/input.txt vendored
View File

0
pytorch_pretrained_bert/tests/fixtures/sample_text.txt → pytorch_transformers/tests/fixtures/sample_text.txt vendored
View File

0
pytorch_pretrained_bert/tests/fixtures/test_sentencepiece.model → pytorch_transformers/tests/fixtures/test_sentencepiece.model vendored
View File

2
pytorch_pretrained_bert/tests/model_tests_commons.py → pytorch_transformers/tests/model_tests_commons.py
View File

@ -412,7 +412,7 @@ class GPTModelTester(object):
[[], []])
def create_and_check_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(self.base_model_class.PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = self.base_model_class.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

4
pytorch_pretrained_bert/tests/model_utils_test.py → pytorch_transformers/tests/model_utils_test.py
View File

@ -25,8 +25,8 @@ import pytest
import torch
from pytorch_pretrained_bert import PretrainedConfig, PreTrainedModel
from pytorch_pretrained_bert.modeling_bert import BertModel, BertConfig, PRETRAINED_MODEL_ARCHIVE_MAP, PRETRAINED_CONFIG_ARCHIVE_MAP
from pytorch_transformers import PretrainedConfig, PreTrainedModel
from pytorch_transformers.modeling_bert import BertModel, BertConfig, PRETRAINED_MODEL_ARCHIVE_MAP, PRETRAINED_CONFIG_ARCHIVE_MAP
class ModelUtilsTest(unittest.TestCase):

6
pytorch_pretrained_bert/tests/modeling_bert_test.py → pytorch_transformers/tests/modeling_bert_test.py
View File

@ -20,11 +20,11 @@ import unittest
import shutil
import pytest
from pytorch_pretrained_bert import (BertConfig, BertModel, BertForMaskedLM,
from pytorch_transformers import (BertConfig, BertModel, BertForMaskedLM,
BertForNextSentencePrediction, BertForPreTraining,
BertForQuestionAnswering, BertForSequenceClassification,
BertForTokenClassification, BertForMultipleChoice)
from pytorch_pretrained_bert.modeling_bert import PRETRAINED_MODEL_ARCHIVE_MAP
from pytorch_transformers.modeling_bert import PRETRAINED_MODEL_ARCHIVE_MAP
from .model_tests_commons import (create_and_check_commons, ConfigTester, ids_tensor)
@ -266,7 +266,7 @@ class BertModelTest(unittest.TestCase):
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = BertModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

2
pytorch_pretrained_bert/tests/modeling_gpt2_test.py → pytorch_transformers/tests/modeling_gpt2_test.py
View File

@ -25,7 +25,7 @@ import pytest
import torch
from pytorch_pretrained_bert import (GPT2Config, GPT2Model,
from pytorch_transformers import (GPT2Config, GPT2Model,
GPT2LMHeadModel, GPT2DoubleHeadsModel)
from .model_tests_commons import (create_and_check_commons, ConfigTester, GPTModelTester)

2
pytorch_pretrained_bert/tests/modeling_openai_test.py → pytorch_transformers/tests/modeling_openai_test.py
View File

@ -21,7 +21,7 @@ import pytest
import torch
from pytorch_pretrained_bert import (OpenAIGPTConfig, OpenAIGPTModel,
from pytorch_transformers import (OpenAIGPTConfig, OpenAIGPTModel,
OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel)
from .model_tests_commons import (create_and_check_commons, ConfigTester, GPTModelTester)

6
pytorch_pretrained_bert/tests/modeling_transfo_xl_test.py → pytorch_transformers/tests/modeling_transfo_xl_test.py
View File

@ -25,8 +25,8 @@ import pytest
import torch
from pytorch_pretrained_bert import (TransfoXLConfig, TransfoXLModel, TransfoXLLMHeadModel)
from pytorch_pretrained_bert.modeling_transfo_xl import PRETRAINED_MODEL_ARCHIVE_MAP
from pytorch_transformers import (TransfoXLConfig, TransfoXLModel, TransfoXLLMHeadModel)
from pytorch_transformers.modeling_transfo_xl import PRETRAINED_MODEL_ARCHIVE_MAP
from .model_tests_commons import ConfigTester, create_and_check_commons, ids_tensor
@ -184,7 +184,7 @@ class TransfoXLModelTest(unittest.TestCase):
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = TransfoXLModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

6
pytorch_pretrained_bert/tests/modeling_xlm_test.py → pytorch_transformers/tests/modeling_xlm_test.py
View File

@ -20,8 +20,8 @@ import unittest
import shutil
import pytest
from pytorch_pretrained_bert import (XLMConfig, XLMModel, XLMWithLMHeadModel, XLMForQuestionAnswering, XLMForSequenceClassification)
from pytorch_pretrained_bert.modeling_xlm import PRETRAINED_MODEL_ARCHIVE_MAP
from pytorch_transformers import (XLMConfig, XLMModel, XLMWithLMHeadModel, XLMForQuestionAnswering, XLMForSequenceClassification)
from pytorch_transformers.modeling_xlm import PRETRAINED_MODEL_ARCHIVE_MAP
from .model_tests_commons import (create_and_check_commons, ConfigTester, ids_tensor)
@ -250,7 +250,7 @@ class XLMModelTest(unittest.TestCase):
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = XLMModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

6
pytorch_pretrained_bert/tests/modeling_xlnet_test.py → pytorch_transformers/tests/modeling_xlnet_test.py
View File

@ -25,8 +25,8 @@ import pytest
import torch
from pytorch_pretrained_bert import (XLNetConfig, XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering)
from pytorch_pretrained_bert.modeling_xlnet import PRETRAINED_MODEL_ARCHIVE_MAP
from pytorch_transformers import (XLNetConfig, XLNetModel, XLNetLMHeadModel, XLNetForSequenceClassification, XLNetForQuestionAnswering)
from pytorch_transformers.modeling_xlnet import PRETRAINED_MODEL_ARCHIVE_MAP
from .model_tests_commons import ConfigTester, create_and_check_commons, ids_tensor
@ -278,7 +278,7 @@ class XLNetModelTest(unittest.TestCase):
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = XLNetModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

6
pytorch_pretrained_bert/tests/optimization_test.py → pytorch_transformers/tests/optimization_test.py
View File

@ -20,9 +20,9 @@ import unittest
import torch
from pytorch_pretrained_bert import BertAdam
from pytorch_pretrained_bert import OpenAIAdam
from pytorch_pretrained_bert.optimization import ConstantLR, WarmupLinearSchedule, WarmupConstantSchedule, \
from pytorch_transformers import BertAdam
from pytorch_transformers import OpenAIAdam
from pytorch_transformers.optimization import ConstantLR, WarmupLinearSchedule, WarmupConstantSchedule, \
WarmupCosineWithWarmupRestartsSchedule, WarmupCosineWithHardRestartsSchedule, WarmupCosineSchedule
import numpy as np

4
pytorch_pretrained_bert/tests/tokenization_bert_test.py → pytorch_transformers/tests/tokenization_bert_test.py
View File

@ -20,7 +20,7 @@ from io import open
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_bert import (BasicTokenizer,
from pytorch_transformers.tokenization_bert import (BasicTokenizer,
BertTokenizer,
WordpieceTokenizer,
_is_control, _is_punctuation,
@ -51,7 +51,7 @@ class TokenizationTest(unittest.TestCase):
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = BertTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

4
pytorch_pretrained_bert/tests/tokenization_gpt2_test.py → pytorch_transformers/tests/tokenization_gpt2_test.py
View File

@ -20,7 +20,7 @@ import json
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_gpt2 import GPT2Tokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from pytorch_transformers.tokenization_gpt2 import GPT2Tokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from .tokenization_tests_commons import create_and_check_tokenizer_commons
@ -58,7 +58,7 @@ class GPT2TokenizationTest(unittest.TestCase):
# @pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = GPT2Tokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

4
pytorch_pretrained_bert/tests/tokenization_openai_test.py → pytorch_transformers/tests/tokenization_openai_test.py
View File

@ -20,7 +20,7 @@ import json
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_openai import OpenAIGPTTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from pytorch_transformers.tokenization_openai import OpenAIGPTTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from.tokenization_tests_commons import create_and_check_tokenizer_commons
@ -60,7 +60,7 @@ class OpenAIGPTTokenizationTest(unittest.TestCase):
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = OpenAIGPTTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

0
pytorch_pretrained_bert/tests/tokenization_tests_commons.py → pytorch_transformers/tests/tokenization_tests_commons.py
View File

4
pytorch_pretrained_bert/tests/tokenization_transfo_xl_test.py → pytorch_transformers/tests/tokenization_transfo_xl_test.py
View File

@ -20,7 +20,7 @@ from io import open
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_transfo_xl import TransfoXLTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from pytorch_transformers.tokenization_transfo_xl import TransfoXLTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from.tokenization_tests_commons import create_and_check_tokenizer_commons
@ -61,7 +61,7 @@ class TransfoXLTokenizationTest(unittest.TestCase):
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = TransfoXLTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

4
pytorch_pretrained_bert/tests/tokenization_xlm_test.py → pytorch_transformers/tests/tokenization_xlm_test.py
View File

@ -20,7 +20,7 @@ import json
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_xlm import XLMTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from pytorch_transformers.tokenization_xlm import XLMTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
from.tokenization_tests_commons import create_and_check_tokenizer_commons
@ -59,7 +59,7 @@ class XLMTokenizationTest(unittest.TestCase):
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = XLMTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

4
pytorch_pretrained_bert/tests/tokenization_xlnet_test.py → pytorch_transformers/tests/tokenization_xlnet_test.py
View File

@ -19,7 +19,7 @@ import unittest
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_xlnet import (XLNetTokenizer,
from pytorch_transformers.tokenization_xlnet import (XLNetTokenizer,
PRETRAINED_VOCAB_ARCHIVE_MAP,
SPIECE_UNDERLINE)
@ -62,7 +62,7 @@ class XLNetTokenizationTest(unittest.TestCase):
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
cache_dir = "/tmp/pytorch_transformers_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = XLNetTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)

0
pytorch_pretrained_bert/tokenization_bert.py → pytorch_transformers/tokenization_bert.py
View File

0
pytorch_pretrained_bert/tokenization_gpt2.py → pytorch_transformers/tokenization_gpt2.py
View File

0
pytorch_pretrained_bert/tokenization_openai.py → pytorch_transformers/tokenization_openai.py
View File

0
pytorch_pretrained_bert/tokenization_transfo_xl.py → pytorch_transformers/tokenization_transfo_xl.py
View File

0
pytorch_pretrained_bert/tokenization_xlm.py → pytorch_transformers/tokenization_xlm.py
View File

0
pytorch_pretrained_bert/tokenization_xlnet.py → pytorch_transformers/tokenization_xlnet.py
View File

10
setup.py
View File

@ -37,16 +37,16 @@ from io import open
from setuptools import find_packages, setup
setup(
name="pytorch_pretrained_bert",
version="0.6.2",
author="Thomas Wolf, Victor Sanh, Tim Rault, Google AI Language Team Authors, Open AI team Authors",
name="pytorch_transformers",
version="0.7.0",
author="Thomas Wolf, Lysandre Debut, Victor Sanh, Tim Rault, Google AI Language Team Authors, Open AI team Authors",
author_email="thomas@huggingface.co",
description="PyTorch version of Google AI BERT model with script to load Google pre-trained models",
long_description=open("README.md", "r", encoding='utf-8').read(),
long_description_content_type="text/markdown",
keywords='BERT NLP deep learning google',
license='Apache',
url="https://github.com/huggingface/pytorch-pretrained-BERT",
url="https://github.com/huggingface/pytorch-transformers",
packages=find_packages(exclude=["*.tests", "*.tests.*",
"tests.*", "tests"]),
install_requires=['torch>=0.4.1',
@ -58,7 +58,7 @@ setup(
'sentencepiece'],
entry_points={
'console_scripts': [
"pytorch_pretrained_bert=pytorch_pretrained_bert.__main__:main",
"pytorch_transformers=pytorch_transformers.__main__:main",
]
},
# python_requires='>=3.5.0',