This PR is the first step of a code refactoring

.travis.yml

@@ -42,7 +42,7 @@ script:
   # 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
   # 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
   # test speech2text translation
@@ -57,9 +57,9 @@ script:
   # 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 -epochs 1 && 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 onmt/tests/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 onmt/tests/test_model.pt
 
 env:
   global:

docs/source/conf.py

@@ -37,8 +37,8 @@
 from recommonmark.transform import AutoStructify
 
 source_parsers = {
-        '.md': CommonMarkParser,
-    }
+    '.md': CommonMarkParser,
+}
 
 source_suffix = ['.rst', '.md']
 extensions = ['sphinx.ext.autodoc',

onmt/__init__.py

@@ -1,11 +1,4 @@
-import onmt.io
-import onmt.Models
-import onmt.Loss
-import onmt.translate
-import onmt.opts
-from onmt.Trainer import Trainer, Statistics
-from onmt.Optim import Optim
+""" Main entry point of the ONMT library """
+from onmt.trainer import Trainer
 
-# For flake8 compatibility
-__all__ = [onmt.Loss, onmt.Models, onmt.opts,
-           Trainer, Optim, Statistics, onmt.io, onmt.translate]
+__all__ = ["Trainer"]

onmt/decoders/__init__.py

@@ -0,0 +1 @@
+"""Module defining decoders."""

onmt/modules/Conv2Conv.py → onmt/decoders/cnn_decoder.py

@@ -1,95 +1,18 @@
 """
-Implementation of "Convolutional Sequence to Sequence Learning"
+Implementation of the CNN Decoder part of
+"Convolutional Sequence to Sequence Learning"
 """
 import torch
 import torch.nn as nn
-import torch.nn.init as init
-import torch.nn.functional as F
 
 import onmt.modules
-from onmt.modules.WeightNorm import WeightNormConv2d
-from onmt.Models import EncoderBase
-from onmt.Models import DecoderState
-from onmt.Utils import aeq
-
+from onmt.decoders.decoder import DecoderState
+from onmt.utils.misc import aeq
+from onmt.utils.cnn_factory import shape_transform, GatedConv
 
 SCALE_WEIGHT = 0.5 ** 0.5
 
 
-def shape_transform(x):
-    """ Tranform the size of the tensors to fit for conv input. """
-    return torch.unsqueeze(torch.transpose(x, 1, 2), 3)
-
-
-class GatedConv(nn.Module):
-    def __init__(self, input_size, width=3, dropout=0.2, nopad=False):
-        super(GatedConv, self).__init__()
-        self.conv = WeightNormConv2d(input_size, 2 * input_size,
-                                     kernel_size=(width, 1), stride=(1, 1),
-                                     padding=(width // 2 * (1 - nopad), 0))
-        init.xavier_uniform_(self.conv.weight, gain=(4 * (1 - dropout))**0.5)
-        self.dropout = nn.Dropout(dropout)
-
-    def forward(self, x_var, hidden=None):
-        x_var = self.dropout(x_var)
-        x_var = self.conv(x_var)
-        out, gate = x_var.split(int(x_var.size(1) / 2), 1)
-        out = out * F.sigmoid(gate)
-        return out
-
-
-class StackedCNN(nn.Module):
-    def __init__(self, num_layers, input_size, cnn_kernel_width=3,
-                 dropout=0.2):
-        super(StackedCNN, self).__init__()
-        self.dropout = dropout
-        self.num_layers = num_layers
-        self.layers = nn.ModuleList()
-        for i in range(num_layers):
-            self.layers.append(
-                GatedConv(input_size, cnn_kernel_width, dropout))
-
-    def forward(self, x, hidden=None):
-        for conv in self.layers:
-            x = x + conv(x)
-            x *= SCALE_WEIGHT
-        return x
-
-
-class CNNEncoder(EncoderBase):
-    """
-    Encoder built on CNN based on
-    :cite:`DBLP:journals/corr/GehringAGYD17`.
-    """
-
-    def __init__(self, num_layers, hidden_size,
-                 cnn_kernel_width, dropout, embeddings):
-        super(CNNEncoder, self).__init__()
-
-        self.embeddings = embeddings
-        input_size = embeddings.embedding_size
-        self.linear = nn.Linear(input_size, hidden_size)
-        self.cnn = StackedCNN(num_layers, hidden_size,
-                              cnn_kernel_width, dropout)
-
-    def forward(self, input, lengths=None, hidden=None):
-        """ See :obj:`onmt.modules.EncoderBase.forward()`"""
-        self._check_args(input, lengths, hidden)
-
-        emb = self.embeddings(input)
-        s_len, batch, emb_dim = emb.size()
-
-        emb = emb.transpose(0, 1).contiguous()
-        emb_reshape = emb.view(emb.size(0) * emb.size(1), -1)
-        emb_remap = self.linear(emb_reshape)
-        emb_remap = emb_remap.view(emb.size(0), emb.size(1), -1)
-        emb_remap = shape_transform(emb_remap)
-        out = self.cnn(emb_remap)
-
-        return emb_remap.squeeze(3).transpose(0, 1).contiguous(),\
-            out.squeeze(3).transpose(0, 1).contiguous()
-
-
 class CNNDecoder(nn.Module):
     """
     Decoder built on CNN, based on :cite:`DBLP:journals/corr/GehringAGYD17`.
@@ -114,13 +37,13 @@ def __init__(self, num_layers, hidden_size, attn_type,
         input_size = self.embeddings.embedding_size
         self.linear = nn.Linear(input_size, self.hidden_size)
         self.conv_layers = nn.ModuleList()
-        for i in range(self.num_layers):
+        for _ in range(self.num_layers):
             self.conv_layers.append(
                 GatedConv(self.hidden_size, self.cnn_kernel_width,
                           self.dropout, True))
 
         self.attn_layers = nn.ModuleList()
-        for i in range(self.num_layers):
+        for _ in range(self.num_layers):
             self.attn_layers.append(
                 onmt.modules.ConvMultiStepAttention(self.hidden_size))
 
@@ -134,10 +57,12 @@ def __init__(self, num_layers, hidden_size, attn_type,
 
     def forward(self, tgt, memory_bank, state, memory_lengths=None):
         """ See :obj:`onmt.modules.RNNDecoderBase.forward()`"""
+        # NOTE: memory_lengths is only here for compatibility reasons
+        #       with onmt.modules.RNNDecoderBase.forward()
         # CHECKS
         assert isinstance(state, CNNDecoderState)
-        tgt_len, tgt_batch, _ = tgt.size()
-        contxt_len, contxt_batch, _ = memory_bank.size()
+        _, tgt_batch, _ = tgt.size()
+        _, contxt_batch, _ = memory_bank.size()
         aeq(tgt_batch, contxt_batch)
         # END CHECKS
 
@@ -194,11 +119,18 @@ def forward(self, tgt, memory_bank, state, memory_lengths=None):
 
         return outputs, state, attns
 
-    def init_decoder_state(self, src, memory_bank, enc_hidden):
+    def init_decoder_state(self, _, memory_bank, enc_hidden):
+        """
+        Init decoder state.
+        """
         return CNNDecoderState(memory_bank, enc_hidden)
 
 
 class CNNDecoderState(DecoderState):
+    """
+    Init CNN decoder state.
+    """
+
     def __init__(self, memory_bank, enc_hidden):
         self.init_src = (memory_bank + enc_hidden) * SCALE_WEIGHT
         self.previous_input = None
@@ -210,9 +142,12 @@ def _all(self):
         """
         return (self.previous_input,)
 
-    def update_state(self, input):
+    def detach(self):
+        self.previous_input = self.previous_input.detach()
+
+    def update_state(self, new_input):
         """ Called for every decoder forward pass. """
-        self.previous_input = input
+        self.previous_input = new_input
 
     def repeat_beam_size_times(self, beam_size):
         """ Repeat beam_size times along batch dimension. """