def forward(self, input):
return F.conv1d(input, self.alpha * Variable(self.delta) + self.beta * normalize(self.weight),
self.bias, self.stride, self.padding, self.dilation)
python类conv1d()的实例源码
def test_dirac_identity(self):
batch, in_c, out_c, size, kernel_size = 8, 3, 4, 5, 3
# Test 1D
input_var = Variable(torch.randn(batch, in_c, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size))
init.dirac(filter_var)
output_var = F.conv1d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data # Variables do not support nonzero
self.assertEqual(input_tensor[:, :, 1:-1], output_tensor[:, :in_c, :]) # Assert in_c outputs are preserved
assert torch.nonzero(output_tensor[:, in_c:, :]).numel() == 0 # Assert extra outputs are 0
# Test 2D
input_var = Variable(torch.randn(batch, in_c, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv2d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :]).numel() == 0
# Test 3D
input_var = Variable(torch.randn(batch, in_c, size, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv3d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :, :]).numel() == 0
def test_dirac_identity(self):
batch, in_c, out_c, size, kernel_size = 8, 3, 4, 5, 3
# Test 1D
input_var = Variable(torch.randn(batch, in_c, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size))
init.dirac(filter_var)
output_var = F.conv1d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data # Variables do not support nonzero
self.assertEqual(input_tensor[:, :, 1:-1], output_tensor[:, :in_c, :]) # Assert in_c outputs are preserved
assert torch.nonzero(output_tensor[:, in_c:, :]).numel() == 0 # Assert extra outputs are 0
# Test 2D
input_var = Variable(torch.randn(batch, in_c, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv2d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :]).numel() == 0
# Test 3D
input_var = Variable(torch.randn(batch, in_c, size, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv3d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :, :]).numel() == 0
def test_calculate_gain_linear(self):
for fn in ['linear', 'conv1d', 'conv2d', 'conv3d', 'conv_transpose2d', 'conv_transpose2d', 'conv_transpose3d']:
gain = init.calculate_gain(fn)
self.assertEqual(gain, 1)
def test_dirac_identity(self):
batch, in_c, out_c, size, kernel_size = 8, 3, 4, 5, 3
# Test 1D
input_var = Variable(torch.randn(batch, in_c, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size))
init.dirac(filter_var)
output_var = F.conv1d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data # Variables do not support nonzero
self.assertEqual(input_tensor[:, :, 1:-1], output_tensor[:, :in_c, :]) # Assert in_c outputs are preserved
assert torch.nonzero(output_tensor[:, in_c:, :]).numel() == 0 # Assert extra outputs are 0
# Test 2D
input_var = Variable(torch.randn(batch, in_c, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv2d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :]).numel() == 0
# Test 3D
input_var = Variable(torch.randn(batch, in_c, size, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv3d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :, :]).numel() == 0
def forward(self, input):
return F.conv1d(input, self.weight, self.bias, self.stride,
self.padding, self.dilation, self.groups)
def _shift(self, wg_vb, shift_vb):
"""
variables needed:
wg_vb: [batch_size x num_heads x mem_hei]
shift_vb: [batch_size x num_heads x num_allowed_shifts]
-> sum=1; the shift weight vector
returns:
ws_vb: [batch_size x num_heads x mem_hei]
-> the attention weight by location focus
"""
batch_size = wg_vb.size(0)
input_dim = wg_vb.size(2); assert input_dim == self.mem_hei
filter_dim = shift_vb.size(2); assert filter_dim == self.num_allowed_shifts
ws_vb = None
for i in range(batch_size): # for each head in each batch, the kernel is different ... seems there's no other way by doing the loop here
for j in range(self.num_heads):
ws_tmp_vb = F.conv1d(wg_vb[i][j].unsqueeze(0).unsqueeze(0).repeat(1, 1, 3),
shift_vb[i][j].unsqueeze(0).unsqueeze(0).contiguous(),
padding=filter_dim//2)[:, :, input_dim:(2*input_dim)]
if ws_vb is None:
ws_vb = ws_tmp_vb
else:
ws_vb = torch.cat((ws_vb, ws_tmp_vb), 0)
ws_vb = ws_vb.view(-1, self.num_heads, self.mem_hei)
return ws_vb
def test_dirac_identity(self):
batch, in_c, out_c, size, kernel_size = 8, 3, 4, 5, 3
# Test 1D
input_var = Variable(torch.randn(batch, in_c, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size))
init.dirac(filter_var)
output_var = F.conv1d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data # Variables do not support nonzero
self.assertEqual(input_tensor[:, :, 1:-1], output_tensor[:, :in_c, :]) # Assert in_c outputs are preserved
assert torch.nonzero(output_tensor[:, in_c:, :]).numel() == 0 # Assert extra outputs are 0
# Test 2D
input_var = Variable(torch.randn(batch, in_c, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv2d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :]).numel() == 0
# Test 3D
input_var = Variable(torch.randn(batch, in_c, size, size, size))
filter_var = Variable(torch.zeros(out_c, in_c, kernel_size, kernel_size, kernel_size))
init.dirac(filter_var)
output_var = F.conv3d(input_var, filter_var)
input_tensor, output_tensor = input_var.data, output_var.data
self.assertEqual(input_tensor[:, :, 1:-1, 1:-1, 1:-1], output_tensor[:, :in_c, :, :])
assert torch.nonzero(output_tensor[:, in_c:, :, :, :]).numel() == 0
def test_calculate_gain_linear(self):
for fn in ['linear', 'conv1d', 'conv2d', 'conv3d', 'conv_transpose2d', 'conv_transpose2d', 'conv_transpose3d']:
gain = init.calculate_gain(fn)
self.assertEqual(gain, 1)
def test_calculate_gain_linear(self):
for fn in ['linear', 'conv1d', 'conv2d', 'conv3d', 'conv_transpose2d', 'conv_transpose2d', 'conv_transpose3d']:
gain = init.calculate_gain(fn)
self.assertEqual(gain, 1)
def forward(self, decoder_input, z, drop_prob):
"""
:param decoder_input: tensor with shape of [batch_size, seq_len, embed_size]
:param z: sequence latent variable with shape of [batch_size, latent_variable_size]
:param drop_prob: probability of an element of decoder input to be zeroed in sense of dropout
:return: unnormalized logits of sentense words distribution probabilities
with shape of [batch_size, seq_len, word_vocab_size]
"""
assert parameters_allocation_check(self), \
'Invalid CUDA options. Parameters should be allocated in the same memory'
[batch_size, seq_len, _] = decoder_input.size()
'''
decoder is conditioned on context via additional bias = W_cond * z to every input token
'''
z = t.cat([z] * seq_len, 1).view(batch_size, seq_len, self.params.latent_variable_size)
decoder_input = t.cat([decoder_input, z], 2)
decoder_input = F.dropout(decoder_input, drop_prob)
# x is tensor with shape [batch_size, input_size=in_channels, seq_len=input_width]
x = decoder_input.transpose(1, 2).contiguous()
for layer, kernel in enumerate(self.kernels):
# apply conv layer with non-linearity and drop last elements of sequence to perfrom input shifting
x = F.conv1d(x, kernel,
bias=self.biases[layer],
dilation=self.params.decoder_dilations[layer],
padding=self.params.decoder_paddings[layer])
x_width = x.size()[2]
x = x[:, :, :(x_width - self.params.decoder_paddings[layer])].contiguous()
x = F.relu(x)
x = x.transpose(1, 2).contiguous()
x = x.view(-1, self.out_size)
x = self.fc(x)
result = x.view(-1, seq_len, self.params.word_vocab_size)
return result
def test_calculate_gain_linear(self):
for fn in ['linear', 'conv1d', 'conv2d', 'conv3d', 'conv_transpose2d', 'conv_transpose2d', 'conv_transpose3d']:
gain = init.calculate_gain(fn)
self.assertEqual(gain, 1)
