Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: IBM/pytorch-seq2seq
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: master
Choose a base ref
...
head repository: IBM/pytorch-seq2seq
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: develop
Choose a head ref
Able to merge. These branches can be automatically merged.
13 contributors

Commits on Sep 28, 2017

  1. Merge branch 'master' into develop

    @kylegao91
    kylegao91 committed Sep 28, 2017
    Copy the full SHA
    437dede View commit details

  2. Modified parameter order of DecoderRNN.forward (#85)

    @liuq901 @kylegao91
    liuq901 authored and kylegao91 committed Sep 28, 2017
    Copy the full SHA
    507849b View commit details

Commits on Oct 1, 2017

  1. Fixed topk decoder.

    @kylegao91
    kylegao91 committed Oct 1, 2017
    Copy the full SHA
    82ed883 View commit details

Commits on Oct 4, 2017

  1. Fixed unit test.

    @kylegao91
    kylegao91 committed Oct 4, 2017
    Copy the full SHA
    57d72c7 View commit details

  2. Make consistent of outputs of Decoder and TopKDecoder.

    @kylegao91
    kylegao91 committed Oct 4, 2017
    Copy the full SHA
    2bfe47c View commit details

  3. Fixed division in python3

    @kylegao91
    kylegao91 committed Oct 4, 2017
    Copy the full SHA
    33cd333 View commit details

  4. Updated apidocs.

    @kylegao91
    kylegao91 committed Oct 4, 2017
    Copy the full SHA
    19f4615 View commit details

  5. Updated code style.

    @kylegao91
    kylegao91 committed Oct 4, 2017
    Copy the full SHA
    c5ccf32 View commit details

Commits on Oct 8, 2017

  1. Refactored decoder from decoder rnn.

    @kylegao91
    kylegao91 committed Oct 8, 2017
    Copy the full SHA
    d78d333 View commit details

Commits on Oct 9, 2017

  1. Added Seq2SeqDataset class.

    @kylegao91
    kylegao91 committed Oct 9, 2017
    Copy the full SHA
    7486875 View commit details

  2. Integration test using new dataset class.

    @kylegao91
    kylegao91 committed Oct 9, 2017
    Copy the full SHA
    c3b70e5 View commit details

Commits on Oct 10, 2017

  1. Added index field for indexing.

    @kylegao91
    kylegao91 committed Oct 10, 2017
    Copy the full SHA
    5665934 View commit details

  2. Added copy decoder.

    @kylegao91
    kylegao91 committed Oct 10, 2017
    Copy the full SHA
    48205a1 View commit details

Commits on Oct 12, 2017

  1. Refactored interfaces to take batch instead of variables.

    @kylegao91
    kylegao91 committed Oct 12, 2017
    Copy the full SHA
    5d6b479 View commit details

  2. Refactored the place of symbol generation.

    @kylegao91
    kylegao91 committed Oct 12, 2017
    Copy the full SHA
    799a922 View commit details

  3. Pass batch through decoder.

    @kylegao91
    kylegao91 committed Oct 12, 2017
    Copy the full SHA
    ab06567 View commit details

Commits on Oct 13, 2017

  1. Added dynamic vocabulary to dataset.

    @kylegao91
    kylegao91 committed Oct 13, 2017
    Copy the full SHA
    3613879 View commit details

Commits on Oct 14, 2017

  1. Merge from topk.

    @kylegao91
    kylegao91 committed Oct 14, 2017
    Copy the full SHA
    a6ac847 View commit details

Commits on Oct 24, 2017

  1. Updated TopKDecoder (#86) 

    * Fixed topk decoder.
    @kylegao91
    kylegao91 authored Oct 24, 2017
    Copy the full SHA
    842d8aa View commit details

  2. Implemented CopyDecoder and refactoring.

    @kylegao91
    kylegao91 committed Oct 24, 2017
    Copy the full SHA
    8ec3900 View commit details

  3. Use torchtext from pipy (#87) 

    * Use torchtext from pipe.

* Fixed torch text sorting order.
    @kylegao91
    kylegao91 authored Oct 24, 2017
    Copy the full SHA
    a32999e View commit details

  4. attention is not required when only using teacher forcing in decoder (#… 

    …90)

* attention is not required when only using teacher forcing in decoder
    @lifengjin @kylegao91
    lifengjin authored and kylegao91 committed Oct 24, 2017
    Copy the full SHA
    3f201b8 View commit details

Commits on Oct 30, 2017

  1. Updated docs and version.

    @kylegao91
    kylegao91 committed Oct 30, 2017
    Copy the full SHA
    96af89d View commit details

  2. Fixed code style.

    @kylegao91
    kylegao91 committed Oct 30, 2017
    Copy the full SHA
    1574e1c View commit details

Commits on Nov 1, 2017

  1. bugfix (#92) 

    Fixed field arguments validation.
    @kylegao91
    Taras Sereda authored and kylegao91 committed Nov 1, 2017
    Copy the full SHA
    feabc36 View commit details

Commits on Nov 7, 2017

  1. Use new dataset class in sample.

    @kylegao91
    kylegao91 committed Nov 7, 2017
    Copy the full SHA
    9bbface View commit details

  2. Fixed argument order.

    @kylegao91
    kylegao91 committed Nov 7, 2017
    Copy the full SHA
    d8fcaef View commit details

  3. Fixed noncontinuous tensor view.

    @kylegao91
    kylegao91 committed Nov 7, 2017
    Copy the full SHA
    6469d9b View commit details

  4. Fixed wrong variable.

    @kylegao91
    kylegao91 committed Nov 7, 2017
    Copy the full SHA
    325b62d View commit details

  5. Merge branch 'master' into copy 

    # Conflicts:
#	seq2seq/models/DecoderRNN.py
#	seq2seq/models/TopKDecoder.py
#	tests/test_topkdecoder.py
    @kylegao91
    kylegao91 committed Nov 7, 2017
    Copy the full SHA
    f851cdc View commit details

  6. Updated integration test script with new dataset class.

    @kylegao91
    kylegao91 committed Nov 7, 2017
    Copy the full SHA
    fd274c2 View commit details

Commits on Nov 14, 2017

  1. Removed initial_lr when resuming optimizer with scheduler. (#95)

    @kylegao91
    kylegao91 authored Nov 14, 2017
    Copy the full SHA
    ed8b90c View commit details

  2. shuffle the training data (#97) 

    * 0.1.5 (#91)

* Modified parameter order of DecoderRNN.forward (#85)

* Updated TopKDecoder (#86)

* Fixed topk decoder.

* Use torchtext from pipy (#87)

* Use torchtext from pipe.

* Fixed torch text sorting order.

* attention is not required when only using teacher forcing in decoder (#90)

* attention is not required when only using teacher forcing in decoder

* Updated docs and version.

* Fixed code style.

* shuffle the training data
    @KaidongYu @kylegao91
    KaidongYu authored and kylegao91 committed Nov 14, 2017
    Copy the full SHA
    6ec61bc View commit details

Commits on Nov 20, 2017

  1. fix example of inflate function in TopKDecoer.py (#98) 

    * fix example of inflate function in TopKDecoer.py
    @Zhujunnan @kylegao91
    Zhujunnan authored and kylegao91 committed Nov 20, 2017
    Copy the full SHA
    bd3537e View commit details

  2. Fix hidden_layer size for one-directional decoder (#99) 

    * Fix hidden_layer size for one-directional decoder

Hidden layer size of the decoder was given `hidden_size * 2 if bidirectional else 1`, resulting in a dimensionality error for non-bidirectional decoders.
Changed `1` to `hidden_size`.
    @dieuwkehupkes @kylegao91
    dieuwkehupkes authored and kylegao91 committed Nov 20, 2017
    Copy the full SHA
    626842c View commit details

  3. Adapt load to allow CPU loading of GPU models (#100) 

    * Adapt load to allow CPU loading of GPU models

Add storage parameter to torch.load to allow loading
models on a CPU that are trained on the GPU, depending
on availability of cuda.
    @dieuwkehupkes @kylegao91
    dieuwkehupkes authored and kylegao91 committed Nov 20, 2017
    Copy the full SHA
    97aca03 View commit details

Commits on Dec 4, 2017

  1. Fix wrong parameter use on DecoderRNN (#103) 

    * Fix wrong parameter use on DecoderRNN
    @alejandrodumas @kylegao91
    alejandrodumas authored and kylegao91 committed Dec 4, 2017
    Copy the full SHA
    a3232b0 View commit details

  2. Merge branch 'master' into develop 

    # Conflicts:
#	seq2seq/models/TopKDecoder.py
#	seq2seq/trainer/supervised_trainer.py
    @kylegao91
    kylegao91 committed Dec 4, 2017
    Copy the full SHA
    fc584e9 View commit details

Commits on Jan 4, 2018

  1. Merge branch 'master' into develop

    @kylegao91
    kylegao91 committed Jan 4, 2018
    Copy the full SHA
    f86e8c7 View commit details

  2. Upgrade to pytorch-0.3.0 (#111) 

    * Upgrade to pytorch-0.3.0

* Use pytorch 3.0 in travis env.
    @kylegao91
    kylegao91 authored Jan 4, 2018
    Copy the full SHA
    fdbc4a7 View commit details

  3. Make sure tensor contiguous when attention's not used. (#112)

    @kylegao91
    kylegao91 authored Jan 4, 2018
    Copy the full SHA
    75a8b76 View commit details

Commits on Jan 18, 2018

  1. Implementing the predict_n method. Using the beam search outputs it r… 

    …eturns several seqs for a given seq (#116)

* Adding a predictor method to return n predicted seqs for a src_seq input
(intended to be used along to Beam Search using TopKDecoder)
    @cbjuan @kylegao91
    cbjuan authored and kylegao91 committed Jan 18, 2018
    Copy the full SHA
    38e7e21 View commit details

Commits on Jan 21, 2018

  1. Checkpoint after batches not epochs (#119)

    @tejaswini @kylegao91
    tejaswini authored and kylegao91 committed Jan 21, 2018
    Copy the full SHA
    cbd8d8b View commit details

Commits on Apr 25, 2018

  1. Merge branch 'master' into develop

    @kylegao91
    kylegao91 committed Apr 25, 2018
    Copy the full SHA
    302c1e8 View commit details

  2. Pytorch 0.4 (#134) 

    * add contiguous call to tensor (#127)

when attention is turned off, pytorch (well, 0.4 at least) gets angry about calling view on a non-contiguous tensor

* Fixed shape documentation (#131)

* Update to pytorch-0.4

* Remove pytorch manual install in travis.
    @kylegao91
    kylegao91 authored Apr 25, 2018
    Copy the full SHA
    8995987 View commit details

  3. Merge remote-tracking branch 'origin/develop' into develop

    @kylegao91
    kylegao91 committed Apr 25, 2018
    Copy the full SHA
    519db31 View commit details

Commits on May 4, 2018

  1. Allow using pre-trained embedding (#135)

    @kylegao91
    kylegao91 authored May 4, 2018
    Copy the full SHA
    0e1d875 View commit details

  2. updated docs

    @kylegao91
    kylegao91 committed May 4, 2018
    Copy the full SHA
    ee7d6bd View commit details

  3. updated README

    @kylegao91
    kylegao91 committed May 4, 2018
    Copy the full SHA
    4799b7c View commit details

  4. Merge branch 'master' into develop

    @kylegao91
    kylegao91 authored May 4, 2018
    Copy the full SHA
    1d6c6e2 View commit details
Showing with 6,384 additions and 2,828 deletions.
  1. +18 −0 .gitignore
  2. +577 −0 .pylintrc
  3. +32 −17 README.md
  4. +6 −5 Vagrantfile
  5. +4 −0 docs/.buildinfo
  6. BIN docs/.doctrees/data.doctree
  7. BIN docs/.doctrees/dataset.doctree
  8. BIN docs/.doctrees/environment.pickle
  9. BIN docs/.doctrees/evaluator.doctree
  10. BIN docs/.doctrees/index.doctree
  11. BIN docs/.doctrees/loss.doctree
  12. BIN docs/.doctrees/models.doctree
  13. BIN docs/.doctrees/notes/intro.doctree
  14. BIN docs/.doctrees/optim.doctree
  15. BIN docs/.doctrees/trainer.doctree
  16. BIN docs/.doctrees/util.doctree
  17. +0 −20 docs/Makefile
  18. +21 −0 docs/README.md
  19. 0 docs/{ → _dev}/requirements.txt
  20. +3 −3 docs/{ → _dev}/source/conf.py
  21. +7 −6 docs/{source/dataset.rst → _dev/source/data.rst}
  22. +2 −4 docs/{ → _dev}/source/evaluator.rst
  23. +1 −1 docs/{ → _dev}/source/index.rst
  24. +20 −0 docs/_dev/source/loss.rst
  25. +45 −0 docs/_dev/source/models.rst
  26. +90 −0 docs/_dev/source/notes/intro.md
  27. +0 −2 docs/{ → _dev}/source/optim.rst
  28. +1 −2 docs/{ → _dev}/source/trainer.rst
  29. +1 −2 docs/{ → _dev}/source/util.rst
  30. +35 −37 docs/{public → }/_modules/index.html
  31. +12 −0 docs/_sources/data.rst.txt
  32. +7 −0 docs/{public → }/_sources/dataset.rst.txt
  33. +2 −4 docs/{public → }/_sources/evaluator.rst.txt
  34. +1 −1 docs/{public → }/_sources/index.rst.txt
  35. +20 −0 docs/_sources/loss.rst.txt
  36. +45 −0 docs/_sources/models.rst.txt
  37. +90 −0 docs/_sources/notes/intro.md.txt
  38. +0 −2 docs/{public → }/_sources/optim.rst.txt
  39. +1 −2 docs/{public → }/_sources/trainer.rst.txt
  40. +1 −2 docs/{public → }/_sources/util.rst.txt
  41. BIN docs/{public → }/_static/ajax-loader.gif
  42. +29 −3 docs/{public → }/_static/basic.css
  43. BIN docs/{public → }/_static/comment-bright.png
  44. BIN docs/{public → }/_static/comment-close.png
  45. BIN docs/{public → }/_static/comment.png
  46. +1 −0 docs/_static/css/badge_only.css
  47. +6 −0 docs/_static/css/theme.css
  48. +43 −17 docs/{public → }/_static/doctools.js
  49. +9 −0 docs/_static/documentation_options.js
  50. BIN docs/{public → }/_static/down-pressed.png
  51. BIN docs/{public → }/_static/down.png
  52. BIN docs/{public → }/_static/file.png
  53. BIN docs/_static/fonts/Lato/lato-bold.eot
  54. BIN docs/_static/fonts/Lato/lato-bold.ttf
  55. BIN docs/_static/fonts/Lato/lato-bold.woff
  56. BIN docs/_static/fonts/Lato/lato-bold.woff2
  57. BIN docs/_static/fonts/Lato/lato-bolditalic.eot
  58. BIN docs/_static/fonts/Lato/lato-bolditalic.ttf
  59. BIN docs/_static/fonts/Lato/lato-bolditalic.woff
  60. BIN docs/_static/fonts/Lato/lato-bolditalic.woff2
  61. BIN docs/_static/fonts/Lato/lato-italic.eot
  62. BIN docs/_static/fonts/Lato/lato-italic.ttf
  63. BIN docs/_static/fonts/Lato/lato-italic.woff
  64. BIN docs/_static/fonts/Lato/lato-italic.woff2
  65. BIN docs/_static/fonts/Lato/lato-regular.eot
  66. BIN docs/_static/fonts/Lato/lato-regular.ttf
  67. BIN docs/_static/fonts/Lato/lato-regular.woff
  68. BIN docs/_static/fonts/Lato/lato-regular.woff2
  69. BIN docs/_static/fonts/RobotoSlab/roboto-slab-v7-bold.eot
  70. BIN docs/{public/_static/fonts/RobotoSlab-Bold.ttf → _static/fonts/RobotoSlab/roboto-slab-v7-bold.ttf}
  71. BIN docs/_static/fonts/RobotoSlab/roboto-slab-v7-bold.woff
  72. BIN docs/_static/fonts/RobotoSlab/roboto-slab-v7-bold.woff2
  73. BIN docs/_static/fonts/RobotoSlab/roboto-slab-v7-regular.eot
  74. BIN ...ublic/_static/fonts/RobotoSlab-Regular.ttf → _static/fonts/RobotoSlab/roboto-slab-v7-regular.ttf}
  75. BIN docs/_static/fonts/RobotoSlab/roboto-slab-v7-regular.woff
  76. BIN docs/_static/fonts/RobotoSlab/roboto-slab-v7-regular.woff2
  77. BIN docs/_static/fonts/fontawesome-webfont.eot
  78. +2,671 −0 docs/_static/fonts/fontawesome-webfont.svg
  79. BIN docs/{public → }/_static/fonts/fontawesome-webfont.ttf
  80. BIN docs/_static/fonts/fontawesome-webfont.woff
  81. BIN docs/_static/fonts/fontawesome-webfont.woff2
  82. +474 −295 docs/{public/_static/jquery-3.1.0.js → _static/jquery-3.2.1.js}
  83. +4 −0 docs/_static/jquery.js
  84. 0 docs/{public → }/_static/js/modernizr.min.js
  85. +3 −0 docs/_static/js/theme.js
  86. BIN docs/{public → }/_static/minus.png
  87. BIN docs/{public → }/_static/plus.png
  88. 0 docs/{public → }/_static/pygments.css
  89. +4 −1 docs/{public → }/_static/searchtools.js
  90. 0 docs/{public → }/_static/underscore-1.3.1.js
  91. 0 docs/{public → }/_static/underscore.js
  92. BIN docs/{public → }/_static/up-pressed.png
  93. BIN docs/{public → }/_static/up.png
  94. +2 −2 docs/{public → }/_static/websupport.js
  95. +290 −0 docs/data.html
  96. +74 −36 docs/{public → }/dataset.html
  97. +35 −89 docs/{public → }/evaluator.html
  98. +82 −79 docs/{public → }/genindex.html
  99. +31 −37 docs/{public → }/index.html
  100. +136 −79 docs/{public → }/loss.html
  101. +194 −127 docs/{public → }/models.html
  102. +82 −72 docs/{public → }/notes/intro.html
  103. BIN docs/objects.inv
  104. +35 −51 docs/{public → }/optim.html
  105. +0 −19 docs/public/_sources/loss.rst.txt
  106. +0 −39 docs/public/_sources/models.rst.txt
  107. +0 −70 docs/public/_sources/notes/intro.md.txt
  108. +0 −2 docs/public/_static/css/badge_only.css
  109. +0 −5 docs/public/_static/css/theme.css
  110. BIN docs/public/_static/fonts/Inconsolata-Bold.ttf
  111. BIN docs/public/_static/fonts/Inconsolata-Regular.ttf
  112. BIN docs/public/_static/fonts/Lato-Bold.ttf
  113. BIN docs/public/_static/fonts/Lato-Regular.ttf
  114. BIN docs/public/_static/fonts/fontawesome-webfont.eot
  115. +0 −685 docs/public/_static/fonts/fontawesome-webfont.svg
  116. BIN docs/public/_static/fonts/fontawesome-webfont.woff
  117. +0 −4 docs/public/_static/jquery.js
  118. +0 −169 docs/public/_static/js/theme.js
  119. +0 −9 docs/public/objects.inv
  120. +0 −1 docs/public/searchindex.js
  121. +41 −43 docs/{public → }/py-modindex.html
  122. +25 −33 docs/{public → }/search.html
  123. +1 −0 docs/searchindex.js
  124. +0 −19 docs/source/loss.rst
  125. +0 −39 docs/source/models.rst
  126. +0 −70 docs/source/notes/intro.md
  127. +36 −51 docs/{public → }/trainer.html
  128. +41 −56 docs/{public → }/util.html
  129. +210 −109 examples/sample.py
  130. +6 −5 requirements.txt
  131. +7 −12 scripts/generate_toy_data.py
  132. +27 −39 scripts/integration_test.py
  133. +10 −6 scripts/integration_test.sh
  134. +1 −0 seq2seq/{dataset → data}/__init__.py
  135. +86 −0 seq2seq/data/dataset.py
  136. 0 seq2seq/{dataset → data}/fields.py
  137. +6 −7 seq2seq/evaluator/evaluator.py
  138. +18 −38 seq2seq/evaluator/predictor.py
  139. +1 −1 seq2seq/loss/__init__.py
  140. +30 −21 seq2seq/loss/loss.py
  141. +4 −3 seq2seq/models/__init__.py
  142. +8 −6 seq2seq/models/{baseRNN.py → base_rnn.py}
  143. +43 −49 seq2seq/models/{DecoderRNN.py → decoder_rnn.py}
  144. +5 −6 seq2seq/models/{attention.py → dot_attention.py}
  145. +7 −8 seq2seq/models/{EncoderRNN.py → encoder_rnn.py}
  146. +13 −8 seq2seq/models/seq2seq.py
  147. +18 −0 seq2seq/models/simple_decoder.py
  148. +32 −20 seq2seq/models/{TopKDecoder.py → top_k_decoder.py}
  149. +132 −119 seq2seq/trainer/supervised_trainer.py
  150. +1 −0 seq2seq/util/__init__.py
  151. +100 −0 tests/data/src.txt
  152. +100 −0 tests/data/tgt.txt
  153. +10 −11 tests/test_checkpoint.py
  154. +62 −0 tests/test_dataset.py
  155. +9 −9 tests/test_decoder_rnn.py
  156. +0 −1 tests/test_encoder_rnn.py
  157. +2 −2 tests/test_evaluator.py
  158. +1 −1 tests/test_fields.py
  159. +28 −20 tests/{test_loss_loss.py → test_loss.py}
  160. +3 −3 tests/test_predictor.py
  161. +3 −3 tests/test_supervised_trainer.py
  162. +10 −9 tests/test_topkdecoder.py
18 changes: 18 additions & 0 deletions .gitignore
View file
Original file line number Diff line number Diff line change
@@ -77,4 +77,22 @@ target/
# Vagrant
.vagrant

# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/

vocab_pickle

# VS Code config
.vscode/*

# local dev folders
data/*
experiment/*

.pylintrc
Loading