flake and tests fixes

making fork changes clearer unit test try to install travis travis travis flake fixes flake fixes change tests remove sru from test (need cuda) travis fixes fix flake
Helsinki-NLP · Jun 11, 2018 · e75bdc0 · e75bdc0
1 parent 2723229
commit e75bdc0
Show file tree

Hide file tree

Showing 19 changed files with 209 additions and 232 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -14,7 +14,7 @@ before_install:
   # Useful for debugging any issues with conda
   - conda info -a
   # freeze the supported pytorch version for consistency
-  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION pytorch=0.3.0 -c soumith
+  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION pytorch=0.4.0 -c soumith
   - source activate test-environment
   # use requirements.txt for dependencies
   - pip install -r requirements.txt
@@ -27,41 +27,42 @@ install:
 
 # Please also add tests to `test/pull_request_chk.sh`.
 script:
-  - wget -O /tmp/im2text.tgz http://lstm.seas.harvard.edu/latex/im2text_small.tgz; tar zxf /tmp/im2text.tgz -C /tmp/; head /tmp/im2text/src-train.txt > /tmp/im2text/src-train-head.txt; head /tmp/im2text/tgt-train.txt > /tmp/im2text/tgt-train-head.txt; head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt
-  - wget -O /tmp/speech.tgz http://lstm.seas.harvard.edu/latex/speech.tgz; tar zxf /tmp/speech.tgz -C /tmp/; head /tmp/speech/src-train.txt > /tmp/speech/src-train-head.txt; head /tmp/speech/tgt-train.txt > /tmp/speech/tgt-train-head.txt; head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt
-  - wget -O /tmp/test_model_speech.pt http://lstm.seas.harvard.edu/latex/test_model_speech.pt
-  - wget -O /tmp/test_model_im2text.pt http://lstm.seas.harvard.edu/latex/test_model_im2text.pt
-  - python -m unittest discover
+   - onmt/tests/test_models.sh lstm cnn transformer
+  #- wget -O /tmp/im2text.tgz http://lstm.seas.harvard.edu/latex/im2text_small.tgz; tar zxf /tmp/im2text.tgz -C /tmp/; head /tmp/im2text/src-train.txt > /tmp/im2text/src-train-head.txt; head /tmp/im2text/tgt-train.txt > /tmp/im2text/tgt-train-head.txt; head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt
+  #- wget -O /tmp/speech.tgz http://lstm.seas.harvard.edu/latex/speech.tgz; tar zxf /tmp/speech.tgz -C /tmp/; head /tmp/speech/src-train.txt > /tmp/speech/src-train-head.txt; head /tmp/speech/tgt-train.txt > /tmp/speech/tgt-train-head.txt; head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt
+  #- wget -O /tmp/test_model_speech.pt http://lstm.seas.harvard.edu/latex/test_model_speech.pt
+  #- wget -O /tmp/test_model_im2text.pt http://lstm.seas.harvard.edu/latex/test_model_im2text.pt
+  #- python -m unittest discover
   # test nmt preprocessing
-  - python preprocess.py -train_src data/src-train.txt -train_tgt data/tgt-train.txt -valid_src data/src-val.txt -valid_tgt data/tgt-val.txt -save_data /tmp/data -src_vocab_size 1000 -tgt_vocab_size 1000 && rm -rf /tmp/data*.pt
+  #- python preprocess.py -train_src data/src-train.txt -train_tgt data/tgt-train.txt -valid_src data/src-val.txt -valid_tgt data/tgt-val.txt -save_data /tmp/data -src_vocab_size 1000 -tgt_vocab_size 1000 && rm -rf /tmp/data*.pt
   # test im2text preprocessing
-  - python preprocess.py -data_type img -src_dir /tmp/im2text/images -train_src /tmp/im2text/src-train.txt -train_tgt /tmp/im2text/tgt-train.txt -valid_src /tmp/im2text/src-val.txt -valid_tgt /tmp/im2text/tgt-val.txt -save_data /tmp/im2text/data && rm -rf /tmp/im2text/data*.pt
+  #- python preprocess.py -data_type img -src_dir /tmp/im2text/images -train_src /tmp/im2text/src-train.txt -train_tgt /tmp/im2text/tgt-train.txt -valid_src /tmp/im2text/src-val.txt -valid_tgt /tmp/im2text/tgt-val.txt -save_data /tmp/im2text/data && rm -rf /tmp/im2text/data*.pt
   # test speech2text preprocessing
-  - python preprocess.py -data_type audio -src_dir /tmp/speech/an4_dataset -train_src /tmp/speech/src-train.txt -train_tgt /tmp/speech/tgt-train.txt -valid_src /tmp/speech/src-val.txt -valid_tgt /tmp/speech/tgt-val.txt -save_data /tmp/speech/data && rm -rf /tmp/speech/data*.pt
+  #- python preprocess.py -data_type audio -src_dir /tmp/speech/an4_dataset -train_src /tmp/speech/src-train.txt -train_tgt /tmp/speech/tgt-train.txt -valid_src /tmp/speech/src-val.txt -valid_tgt /tmp/speech/tgt-val.txt -save_data /tmp/speech/data && rm -rf /tmp/speech/data*.pt
   # test nmt translation
-  - head data/src-test.txt > /tmp/src-test.txt; python translate.py -model test/test_model.pt -src /tmp/src-test.txt -verbose
+  #- head data/src-test.txt > /tmp/src-test.txt; python translate.py -model onmt/tests/test_model.pt -src /tmp/src-test.txt -verbose
   # test im2text translation
-  - head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt; python translate.py -data_type img -src_dir /tmp/im2text/images -model /tmp/test_model_im2text.pt -src /tmp/im2text/src-val-head.txt -tgt /tmp/im2text/tgt-val-head.txt -verbose -out /tmp/im2text/trans
+  #- head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt; python translate.py -data_type img -src_dir /tmp/im2text/images -model /tmp/test_model_im2text.pt -src /tmp/im2text/src-val-head.txt -tgt /tmp/im2text/tgt-val-head.txt -verbose -out /tmp/im2text/trans
   # test speech2text translation
-  - head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt; python translate.py -data_type audio -src_dir /tmp/speech/an4_dataset -model /tmp/test_model_speech.pt -src /tmp/speech/src-val-head.txt -tgt /tmp/speech/tgt-val-head.txt -verbose -out /tmp/speech/trans; diff /tmp/speech/tgt-val-head.txt /tmp/speech/trans
+  #- head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt; python translate.py -data_type audio -src_dir /tmp/speech/an4_dataset -model /tmp/test_model_speech.pt -src /tmp/speech/src-val-head.txt -tgt /tmp/speech/tgt-val-head.txt -verbose -out /tmp/speech/trans; diff /tmp/speech/tgt-val-head.txt /tmp/speech/trans
   # test nmt preprocessing and training
-  - head data/src-val.txt > /tmp/src-val.txt; head data/tgt-val.txt > /tmp/tgt-val.txt; python preprocess.py -train_src /tmp/src-val.txt -train_tgt /tmp/tgt-val.txt -valid_src /tmp/src-val.txt -valid_tgt /tmp/tgt-val.txt -save_data /tmp/q -src_vocab_size 1000 -tgt_vocab_size 1000; python train.py -data /tmp/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -train_steps 1000 && rm -rf /tmp/q*.pt
+  #- head data/src-val.txt > /tmp/src-val.txt; head data/tgt-val.txt > /tmp/tgt-val.txt; python preprocess.py -train_src /tmp/src-val.txt -train_tgt /tmp/tgt-val.txt -valid_src /tmp/src-val.txt -valid_tgt /tmp/tgt-val.txt -save_data /tmp/q -src_vocab_size 1000 -tgt_vocab_size 1000; python train.py -data /tmp/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 50 -rnn_size 10 -train_steps 500 && rm -rf /tmp/q*.pt
   # test nmt preprocessing w/ sharding and training w/copy
-  - head data/src-val.txt > /tmp/src-val.txt; head data/tgt-val.txt > /tmp/tgt-val.txt; python preprocess.py -train_src /tmp/src-val.txt -train_tgt /tmp/tgt-val.txt -valid_src /tmp/src-val.txt -valid_tgt /tmp/tgt-val.txt -max_shard_size 1 -dynamic_dict -save_data /tmp/q -src_vocab_size 1000 -tgt_vocab_size 1000; python train.py -data /tmp/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -copy_attn -train_steps 1000 && rm -rf /tmp/q*.pt
+  #- head data/src-val.txt > /tmp/src-val.txt; head data/tgt-val.txt > /tmp/tgt-val.txt; python preprocess.py -train_src /tmp/src-val.txt -train_tgt /tmp/tgt-val.txt -valid_src /tmp/src-val.txt -valid_tgt /tmp/tgt-val.txt -max_shard_size 1 -dynamic_dict -save_data /tmp/q -src_vocab_size 1000 -tgt_vocab_size 1000; python train.py -data /tmp/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 50 -rnn_size 10 -copy_attn -train_steps 500 && rm -rf /tmp/q*.pt
 
   # test im2text preprocessing and training
-  - head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt; python preprocess.py -data_type img -src_dir /tmp/im2text/images -train_src /tmp/im2text/src-val-head.txt -train_tgt /tmp/im2text/tgt-val-head.txt -valid_src /tmp/im2text/src-val-head.txt -valid_tgt /tmp/im2text/tgt-val-head.txt -save_data /tmp/im2text/q; python train.py -model_type img -data /tmp/im2text/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -train_steps 1000 && rm -rf /tmp/im2text/q*.pt
+  #- head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt; python preprocess.py -data_type img -src_dir /tmp/im2text/images -train_src /tmp/im2text/src-val-head.txt -train_tgt /tmp/im2text/tgt-val-head.txt -valid_src /tmp/im2text/src-val-head.txt -valid_tgt /tmp/im2text/tgt-val-head.txt -save_data /tmp/im2text/q; python train.py -model_type img -data /tmp/im2text/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -train_steps 1000 && rm -rf /tmp/im2text/q*.pt
   # test speech2text preprocessing and training
-  - head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt; python preprocess.py -data_type audio -src_dir /tmp/speech/an4_dataset -train_src /tmp/speech/src-val-head.txt -train_tgt /tmp/speech/tgt-val-head.txt -valid_src /tmp/speech/src-val-head.txt -valid_tgt /tmp/speech/tgt-val-head.txt -save_data /tmp/speech/q; python train.py -model_type audio -data /tmp/speech/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -train_steps 1000 && rm -rf /tmp/speech/q*.pt
+  #- head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt; python preprocess.py -data_type audio -src_dir /tmp/speech/an4_dataset -train_src /tmp/speech/src-val-head.txt -train_tgt /tmp/speech/tgt-val-head.txt -valid_src /tmp/speech/src-val-head.txt -valid_tgt /tmp/speech/tgt-val-head.txt -save_data /tmp/speech/q; python train.py -model_type audio -data /tmp/speech/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -train_steps 1000 && rm -rf /tmp/speech/q*.pt
   # test nmt translation
-  - python translate.py -model test/test_model2.pt  -src  data/morph/src.valid  -verbose -batch_size 10 -beam_size 10 -tgt data/morph/tgt.valid -out /tmp/trans; diff  data/morph/tgt.valid /tmp/trans
+  #- python translate.py -model test/test_model2.pt  -src  data/morph/src.valid  -verbose -batch_size 10 -beam_size 10 -tgt data/morph/tgt.valid -out /tmp/trans; diff  data/morph/tgt.valid /tmp/trans
   # test tool
-  - PYTHONPATH=$PYTHONPATH:. python tools/extract_embeddings.py -model test/test_model.pt
+  #- PYTHONPATH=$PYTHONPATH:. python tools/extract_embeddings.py -model test/test_model.pt
 
 env:
   global:
     # Doctr deploy key for OpenNMT/OpenNMT-py
-    - secure: "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"
+    #- secure: "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"
 
 matrix:
   include:
@@ -77,4 +78,4 @@ matrix:
         - pip install -r docs/requirements.txt
         - cd docs/; make html; cd ..
         - set -e
-        - doctr deploy --built-docs docs/build/html/ .
+        #- doctr deploy --built-docs docs/build/html/ .
diff --git a/README.md b/README.md
@@ -1,37 +1,15 @@
 # OpenNMT-py: Open-Source Neural Machine Translation
 
-[![Build Status](https://travis-ci.org/OpenNMT/OpenNMT-py.svg?branch=master)](https://travis-ci.org/OpenNMT/OpenNMT-py)
+[![Build Status](https://travis-ci.org/Ubiqus/OpenNMT-py.svg?branch=master)](https://travis-ci.org/Ubiqus/OpenNMT-py)
 
 
-[OpenNMT](https://opennmt.net) is an open-source (MIT) neural machine translation system which has 3 different implementations.
+This is a fork of OpenNMT-py
+Multi-GPU supported with Torch Distributed (pytorch 0.4)
 
-The genuine one was a Lua version based on the Harvard Seq2Seq framework. [OpenNMT-Lua](https://github.com/OpenNMT/OpenNMT)
+See major changes here: https://github.com/Ubiqus/OpenNMT-py/releases
 
-The [Pytorch](https://github.com/pytorch/pytorch) version is this repo.
+Script for upgrading existing pytorch 0.3 models: tools/03to04.py
 
-The tensorflow version: [OpenNMT-tf](https://github.com/OpenNMT/OpenNMT-tf)
-
-
-OpenNMT-py is designed to be research friendly to try out new ideas in translation, summary, image-to-text, morphology, and many other domains but also ready for production with a full REST API.
-
-Codebase is relatively stable, but PyTorch is still evolving. We currently recommend forking if you need to have stable code.
-
-OpenNMT-py is run as a collaborative open-source project. The original code was written by [Adam Lerer](http://github.com/adamlerer) (NYC) and [Bryan McCann](https://github.com/bmccann).
-Major contributions have come from [Sasha Rush](http://github.com/srush) and his group (Cambridge, MA), [Ben Peters](http://github.com/bpopeters) (Saarbrücken), [Jianyu Zhan](http://github.com/jianyuzhan) (Shenzhen), [Paul Tardy](https://github.com/pltrdy) , [Vincent Nguyen](https://github.com/vince62s) and many others.  
-
-We love contributions. Please consult the Issues page for any [Contributions Welcome](https://github.com/OpenNMT/OpenNMT-py/issues?q=is%3Aissue+is%3Aopen+label%3A%22contributions+welcome%22) tagged post. 
-
-<center style="padding: 40px"><img width="70%" src="http://opennmt.github.io/simple-attn.png" /></center>
-
-
-Table of Contents
-=================
-  * [Full Documentation](http://opennmt.net/OpenNMT-py/)
-  * [Requirements](#requirements)
-  * [Features](#features)
-  * [Quickstart](#quickstart)
-  * [Citation](#citation)
-
 ## Requirements
   python 3, torch >=0.4.0, torchtext >=0.2.3, six, tqdm, future, cupy pynvrtc for SRU
 
@@ -120,29 +98,4 @@ Now you have a model which you can use to predict on new data. We do this by run
 
 Go to tutorial: [How to use GloVe pre-trained embeddings in OpenNMT-py](http://forum.opennmt.net/t/how-to-use-glove-pre-trained-embeddings-in-opennmt-py/1011)
 
-## Pretrained Models
 
-The following pretrained models can be downloaded and used with translate.py.
-
-http://opennmt.net/Models-py/
-
-
-
-## Citation
-
-[OpenNMT technical report](https://doi.org/10.18653/v1/P17-4012)
-
-```
-@inproceedings{opennmt,
-  author    = {Guillaume Klein and
-               Yoon Kim and
-               Yuntian Deng and
-               Jean Senellart and
-               Alexander M. Rush},
-  title     = {OpenNMT: Open-Source Toolkit for Neural Machine Translation},
-  booktitle = {Proc. ACL},
-  year      = {2017},
-  url       = {https://doi.org/10.18653/v1/P17-4012},
-  doi       = {10.18653/v1/P17-4012}
-}
-```
diff --git a/onmt/__init__.py b/onmt/__init__.py
@@ -9,4 +9,8 @@
 import onmt.modules
 from onmt.trainer import Trainer
 
+# For Flake
+__all__ = [onmt.inputters, onmt.encoders, onmt.decoders, onmt.models,
+           onmt.utils, onmt.modules, "Trainer"]
+
 __version__ = "0.4.0"
diff --git a/onmt/decoders/transformer.py b/onmt/decoders/transformer.py
@@ -81,10 +81,10 @@ def forward(self, inputs, memory_bank, src_pad_mask, tgt_pad_mask,
 
         if self.self_attn_type == "scaled-dot":
             query, attn = self.self_attn(all_input, all_input, input_norm,
-                                             mask=dec_mask)
+                                         mask=dec_mask)
         elif self.self_attn_type == "average":
-            query, attn = self.self_attn(input_norm,
-                                             mask=dec_mask, layer_cache=layer_cache, step=step)
+            query, attn = self.self_attn(input_norm, mask=dec_mask,
+                                         layer_cache=layer_cache, step=step)
 
         query = self.drop(query) + inputs
 
@@ -154,7 +154,8 @@ def __init__(self, num_layers, hidden_size, attn_type,
 
         # Build TransformerDecoder.
         self.transformer_layers = nn.ModuleList(
-            [TransformerDecoderLayer(hidden_size, dropout, self_attn_type=self_attn_type)
+            [TransformerDecoderLayer(hidden_size, dropout,
+             self_attn_type=self_attn_type)
              for _ in range(num_layers)])
 
         # TransformerDecoder has its own attention mechanism.
@@ -166,7 +167,6 @@ def __init__(self, num_layers, hidden_size, attn_type,
             self._copy = True
         self.layer_norm = onmt.modules.LayerNorm(hidden_size)
 
-
     def _init_cache(self, memory_bank, memory_lengths=None):
         cache = {}
         batch_size = memory_bank.size(1)
@@ -176,7 +176,8 @@ def _init_cache(self, memory_bank, memory_lengths=None):
             cache["layer_{}".format(l)] = layer_cache
         return cache
 
-    def forward(self, tgt, memory_bank, state, memory_lengths=None, step=None, cache=None):
+    def forward(self, tgt, memory_bank, state, memory_lengths=None,
+                step=None, cache=None):
         """
         See :obj:`onmt.modules.RNNDecoderBase.forward()`
         """
@@ -229,7 +230,9 @@ def forward(self, tgt, memory_bank, state, memory_lengths=None, step=None, cache
                 = self.transformer_layers[i](output, src_memory_bank,
                                              src_pad_mask, tgt_pad_mask,
                                              previous_input=prev_layer_input,
-                                             layer_cache=cache["layer_{}".format(i)] if cache is not None else None,
+                                             layer_cache=cache["layer_{}".
+                                                               format(i)]
+                                             if cache is not None else None,
                                              step=step)
             saved_inputs.append(all_input)
 

diff --git a/onmt/inputters/inputter.py b/onmt/inputters/inputter.py
@@ -457,7 +457,7 @@ def _next_dataset_iterator(self, dataset_iter):
 def build_dataset_iter(datasets, fields, opt, is_train=True):
     """
     This returns user-defined train/validate data iterator for the trainer
-    to iterate over. We implement simple ordered iterator strategy here, 
+    to iterate over. We implement simple ordered iterator strategy here,
     but more sophisticated strategy like curriculum learning is ok too.
     """
     batch_size = opt.batch_size if is_train else opt.valid_batch_size

diff --git a/onmt/models/model_saver.py b/onmt/models/model_saver.py
@@ -76,7 +76,7 @@ def _rm_checkpoint(self, name):
         Remove a checkpoint
 
         Args:
-            name(str): name that indentifies the checkpoint 
+            name(str): name that indentifies the checkpoint
                 (it may be a filepath)
         """
         raise NotImplementedError()