mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-12-11 06:55:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix batch_size position bug in layer_skip

Browse Source
This commit is contained in:
Daniel Povey 2023-02-16 15:13:06 +08:00
parent 686e7e8828
commit e0b8a0cfd0
1 changed files with 6 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
egs/librispeech/ASR/pruned_transducer_stateless7/zipformer.py
View File

@ -401,11 +401,13 @@ class Zipformer(EncoderInterface):
# stacks. The layer_skip_dropout_prob is to discourage it from
# completely ignoring the middle layers, especially early in
# training,
batch_size = x.shape[0]
batch_size = x.shape[1]
skip_x = self.skip_modules[i](outputs[self.skip_layers[i]], x)
if self.training:
mask = (torch.rand((batch_size, 1, 1), device=x.device) >
float(self.layer_skip_dropout_prob))
layer_skip_dropout_prob = float(self.layer_skip_dropout_prob)
if self.training and layer_skip_dropout_prob > 0:
mask = (torch.rand((1, batch_size, 1), device=x.device) >
layer_skip_dropout_prob)
x = torch.where(mask, skip_x, x)
else:
x = skip_x