Release 0.1.6 (#137)

* 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.

* bugfix (#92)

Fixed field arguments validation.

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

* 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

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

* fix example of inflate function in TopKDecoer.py

* 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`.

* 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.

* Fix wrong parameter use on DecoderRNN (#103)

* Fix wrong parameter use on DecoderRNN

* Upgrade to pytorch-0.3.0 (#111)

* Upgrade to pytorch-0.3.0

* Use pytorch 3.0 in travis env.

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

* Implementing the predict_n method. Using the beam search outputs it returns 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)

* Checkpoint after batches not epochs (#119)

* 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.

* Allow using pre-trained embedding (#135)

* updated docs

Author: kylegao91

.travis.yml

@@ -10,8 +10,6 @@ python:
 install:
   - pip install -U pip
   - pip -q install -r requirements.txt
-  - pip -q install "http://download.pytorch.org/whl/cu75/torch-0.2.0.post1-cp27-cp27mu-manylinux1_x86_64.whl; python_version == '2.7'"
-  - pip -q install "http://download.pytorch.org/whl/cu75/torch-0.2.0.post1-cp36-cp36m-manylinux1_x86_64.whl; python_version == '3.6'"
 
   # dev dependencies
   - pip install flake8
@@ -32,4 +30,4 @@ script:
   # Unit test
   - nosetests --with-coverage --cover-erase --cover-package=seq2seq
   # Integration test
-  - "if [[ $TRAVIS_BRANCH =~ (master|develop) ]]; then python setup.py install && scripts/integration_test.sh; fi"
+  - "if [[ $TRAVIS_BRANCH =~ (master|develop) ]]; then python setup.py install && scripts/integration_test.sh; fi"

docs/public/_modules/index.html

@@ -8,7 +8,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>Overview: module code &mdash; pytorch-seq2seq 0.1.5 documentation</title>
+  <title>Overview: module code &mdash; pytorch-seq2seq 0.1.6 documentation</title>
 
 
 
@@ -35,7 +35,7 @@
         <link rel="index" title="Index"
               href="../genindex.html"/>
         <link rel="search" title="Search" href="../search.html"/>
-    <link rel="top" title="pytorch-seq2seq 0.1.5 documentation" href="../index.html"/> 
+    <link rel="top" title="pytorch-seq2seq 0.1.6 documentation" href="../index.html"/> 
 
 
   <script src="../_static/js/modernizr.min.js"></script>
@@ -64,7 +64,7 @@
 
 
               <div class="version">
-                0.1.5
+                0.1.6
               </div>
 
 
@@ -164,8 +164,6 @@
 
   <h1>All modules for which code is available</h1>
 <ul><li><a href="seq2seq/dataset/fields.html">seq2seq.dataset.fields</a></li>
-<li><a href="seq2seq/dataset/utils.html">seq2seq.dataset.utils</a></li>
-<li><a href="seq2seq/dataset/vocabulary.html">seq2seq.dataset.vocabulary</a></li>
 <li><a href="seq2seq/evaluator/evaluator.html">seq2seq.evaluator.evaluator</a></li>
 <li><a href="seq2seq/evaluator/predictor.html">seq2seq.evaluator.predictor</a></li>
 <li><a href="seq2seq/loss/loss.html">seq2seq.loss.loss</a></li>
@@ -178,7 +176,6 @@ <h1>All modules for which code is available</h1>
 <li><a href="seq2seq/optim/optim.html">seq2seq.optim.optim</a></li>
 <li><a href="seq2seq/trainer/supervised_trainer.html">seq2seq.trainer.supervised_trainer</a></li>
 <li><a href="seq2seq/util/checkpoint.html">seq2seq.util.checkpoint</a></li>
-<li><a href="seq2seq/util/custom_time.html">seq2seq.util.custom_time</a></li>
 </ul>
 
            </div>
@@ -215,7 +212,7 @@ <h1>All modules for which code is available</h1>
     <script type="text/javascript">
         var DOCUMENTATION_OPTIONS = {
             URL_ROOT:'../',
-            VERSION:'0.1.5',
+            VERSION:'0.1.6',
             COLLAPSE_INDEX:false,
             FILE_SUFFIX:'.html',
             HAS_SOURCE:  true,