def forward(self,
input,
lengths=None,
hidden=None,
dump_layers=False,
intervention=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)
if dump_layers:
dumped_layers, out = self.cnn(emb_remap, True, intervention)
else:
out = self.cnn(emb_remap, False, intervention)
if dump_layers:
return emb_remap.squeeze(3).transpose(0, 1).contiguous(), \
dumped_layers, \
out.squeeze(3).transpose(0, 1).contiguous()
else:
return emb_remap.squeeze(3).transpose(0, 1).contiguous(), \
out.squeeze(3).transpose(0, 1).contiguous()
def forward(self, tgt, memory_bank, memory_lengths=None, step=None):
""" See :obj:`onmt.modules.RNNDecoderBase.forward()`"""
# NOTE: memory_lengths is only here for compatibility reasons
# with onmt.modules.RNNDecoderBase.forward()
if self.state["previous_input"] is not None:
tgt = torch.cat([self.state["previous_input"], tgt], 0)
# Initialize return variables.
dec_outs = []
attns = {"std": []}
assert not self._copy, "Copy mechanism not yet tested in conv2conv"
if self._copy:
attns["copy"] = []
emb = self.embeddings(tgt)
assert emb.dim() == 3 # len x batch x embedding_dim
tgt_emb = emb.transpose(0, 1).contiguous()
# The output of CNNEncoder.
src_memory_bank_t = memory_bank.transpose(0, 1).contiguous()
# The combination of output of CNNEncoder and source embeddings.
src_memory_bank_c = self.state["src"].transpose(0, 1).contiguous()
# Run the forward pass of the CNNDecoder.
emb_reshape = tgt_emb.contiguous().view(
tgt_emb.size(0) * tgt_emb.size(1), -1)
linear_out = self.linear(emb_reshape)
x = linear_out.view(tgt_emb.size(0), tgt_emb.size(1), -1)
x = shape_transform(x)
pad = torch.zeros(x.size(0), x.size(1),
self.cnn_kernel_width - 1, 1)
pad = pad.type_as(x)
base_target_emb = x
for conv, attention in zip(self.conv_layers, self.attn_layers):
new_target_input = torch.cat([pad, x], 2)
out = conv(new_target_input)
c, attn = attention(base_target_emb, out,
src_memory_bank_t, src_memory_bank_c)
x = (x + (c + out) * SCALE_WEIGHT) * SCALE_WEIGHT
output = x.squeeze(3).transpose(1, 2)
# Process the result and update the attentions.
dec_outs = output.transpose(0, 1).contiguous()
if self.state["previous_input"] is not None:
dec_outs = dec_outs[self.state["previous_input"].size(0):]
attn = attn[:, self.state["previous_input"].size(0):].squeeze()
attn = torch.stack([attn])
attns["std"] = attn
if self._copy:
attns["copy"] = attn
# Update the state.
self.update_state(tgt)
# TODO change the way attns is returned dict => list or tuple (onnx)
return dec_outs, attns
def forward(self, tgt, memory_bank, step=None, **kwargs):
""" See :obj:`onmt.modules.RNNDecoderBase.forward()`"""
if self.state["previous_input"] is not None:
tgt = torch.cat([self.state["previous_input"], tgt], 0)
dec_outs = []
attns = {"std": []}
if self.copy_attn is not None:
attns["copy"] = []
emb = self.embeddings(tgt)
assert emb.dim() == 3 # len x batch x embedding_dim
tgt_emb = emb.transpose(0, 1).contiguous()
# The output of CNNEncoder.
src_memory_bank_t = memory_bank.transpose(0, 1).contiguous()
# The combination of output of CNNEncoder and source embeddings.
src_memory_bank_c = self.state["src"].transpose(0, 1).contiguous()
emb_reshape = tgt_emb.contiguous().view(
tgt_emb.size(0) * tgt_emb.size(1), -1)
linear_out = self.linear(emb_reshape)
x = linear_out.view(tgt_emb.size(0), tgt_emb.size(1), -1)
x = shape_transform(x)
pad = torch.zeros(x.size(0), x.size(1), self.cnn_kernel_width - 1, 1)
pad = pad.type_as(x)
base_target_emb = x
for conv, attention in zip(self.conv_layers, self.attn_layers):
new_target_input = torch.cat([pad, x], 2)
out = conv(new_target_input)
c, attn = attention(base_target_emb, out,
src_memory_bank_t, src_memory_bank_c)
x = (x + (c + out) * SCALE_WEIGHT) * SCALE_WEIGHT
output = x.squeeze(3).transpose(1, 2)
# Process the result and update the attentions.
dec_outs = output.transpose(0, 1).contiguous()
if self.state["previous_input"] is not None:
dec_outs = dec_outs[self.state["previous_input"].size(0):]
attn = attn[:, self.state["previous_input"].size(0):].squeeze()
attn = torch.stack([attn])
attns["std"] = attn
if self.copy_attn is not None:
attns["copy"] = attn
# Update the state.
self.state["previous_input"] = tgt
# TODO change the way attns is returned dict => list or tuple (onnx)
return dec_outs, attns
def forward(self, input, lengths=None, hidden=None):
""" See :obj:`onmt.modules.EncoderBase.forward()`"""
self._check_args(input, lengths, hidden)
emb = self.embeddings(input)
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(), lengths
def forward(self, input, lengths=None, hidden=None):
"""See :class:`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(), lengths
def cnnforward(self, input, lengths=None, hidden=None):
"""See :class:`onmt.modules.EncoderBase.forward()`"""
import ipdb
ipdb.set_trace()
self._check_args(input, lengths, hidden)
emb = self.embeddings(input) #[src_len, bsz, emb_dim]
emb = emb.transpose(0, 1).contiguous() #[bsz, src_len, emb_dim]
emb_reshape = emb.view(emb.size(0) * emb.size(1),
-1) #[(bsz*src_len), emb_dim]
emb_remap = self.linear(emb_reshape) #[(bsz*src_len), emb_dim]
emb_remap = emb_remap.view(emb.size(0), emb.size(1),
-1) #[bsz, src_len, emb_dim]
emb_remap = shape_transform(emb_remap) #[bsz, emb_dim, src_len, 1]
out = self.cnn(emb_remap) #[bsz, emb_dim, src_len, 1]
return emb_remap.squeeze(3).transpose(0, 1).contiguous(), \
out.squeeze(3).transpose(0, 1).contiguous(), lengths #[emb_dim,bsz,src_len],[emb_dim,bsz,src_len], [bsz]
def forward(self, input, lengths=None, hidden=None):
""" See :obj:`onmt.modules.EncoderBase.forward()`"""
self._check_args(input, lengths, hidden)
if self.embeddings is not None:
emb = self.embeddings(input)
else:
emb = input.transpose(0, 1) #torch.Size([26, 64, 500])
# 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) #torch.Size([64, 512, 26, 1])
return emb_remap.squeeze(3).transpose(0, 1).contiguous(), \
out.squeeze(3).transpose(0, 1).contiguous(), lengths
