coqui-ai
/
TTS
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

testing seq_len_norm

pull/10/head
root 2020-01-27 16:02:34 +01:00 committed by erogol
parent 0d17019d22
commit ca33336ae0
2 changed files with 38 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
layers/losses.py
View File

@ -35,7 +35,7 @@ class L1LossMasked(nn.Module):
mask = mask.expand_as(x)
loss = functional.l1_loss(
x * mask, target * mask, reduction='none')
loss = loss.mul(out_weights.cuda()).sum()
loss = loss.mul(out_weights.to(loss.device)).sum()
else:
mask = mask.expand_as(x)
loss = functional.l1_loss(
@ -74,7 +74,7 @@ class MSELossMasked(nn.Module):
mask = mask.expand_as(x)
loss = functional.mse_loss(
x * mask, target * mask, reduction='none')
loss = loss.mul(out_weights.cuda()).sum()
loss = loss.mul(out_weights.to(loss.device)).sum()
else:
mask = mask.expand_as(x)
loss = functional.mse_loss(

39
tests/test_layers.py
View File

@ -131,7 +131,7 @@ class L1LossMaskedTests(unittest.TestCase):
dummy_target = T.zeros(4, 8, 128).float()
dummy_length = (T.ones(4) * 8).long()
output = layer(dummy_input, dummy_target, dummy_length)
assert output.item() == 1.0, "1.0 vs {}".format(output.data[0])
assert output.item() == 1.0, "1.0 vs {}".format(output.item())
# test if padded values of input makes any difference
dummy_input = T.ones(4, 8, 128).float()
@ -140,7 +140,7 @@ class L1LossMaskedTests(unittest.TestCase):
mask = (
(sequence_mask(dummy_length).float() - 1.0) * 100.0).unsqueeze(2)
output = layer(dummy_input + mask, dummy_target, dummy_length)
assert output.item() == 1.0, "1.0 vs {}".format(output.data[0])
assert output.item() == 1.0, "1.0 vs {}".format(output.item())
dummy_input = T.rand(4, 8, 128).float()
dummy_target = dummy_input.detach()
@ -148,4 +148,37 @@ class L1LossMaskedTests(unittest.TestCase):
mask = (
(sequence_mask(dummy_length).float() - 1.0) * 100.0).unsqueeze(2)
output = layer(dummy_input + mask, dummy_target, dummy_length)
assert output.item() == 0, "0 vs {}".format(output.data[0])
assert output.item() == 0, "0 vs {}".format(output.item())
# seq_len_norm = True
# test input == target
layer = L1LossMasked(seq_len_norm=True)
dummy_input = T.ones(4, 8, 128).float()
dummy_target = T.ones(4, 8, 128).float()
dummy_length = (T.ones(4) * 8).long()
output = layer(dummy_input, dummy_target, dummy_length)
assert output.item() == 0.0
# test input != target
dummy_input = T.ones(4, 8, 128).float()
dummy_target = T.zeros(4, 8, 128).float()
dummy_length = (T.ones(4) * 8).long()
output = layer(dummy_input, dummy_target, dummy_length)
assert output.item() == 1.0, "1.0 vs {}".format(output.item())
# test if padded values of input makes any difference
dummy_input = T.ones(4, 8, 128).float()
dummy_target = T.zeros(4, 8, 128).float()
dummy_length = (T.arange(5, 9)).long()
mask = (
(sequence_mask(dummy_length).float() - 1.0) * 100.0).unsqueeze(2)
output = layer(dummy_input + mask, dummy_target, dummy_length)
assert abs(output.item() - 1.0) < 1e-5, "1.0 vs {}".format(output.item())
dummy_input = T.rand(4, 8, 128).float()
dummy_target = dummy_input.detach()
dummy_length = (T.arange(5, 9)).long()
mask = (
(sequence_mask(dummy_length).float() - 1.0) * 100.0).unsqueeze(2)
output = layer(dummy_input + mask, dummy_target, dummy_length)
assert output.item() == 0, "0 vs {}".format(output.item())