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

Reduce cutoff from 100 to 5 for estimating OOM with warmup

Browse Source
This commit is contained in:
Daniel Povey 2022-10-25 12:53:12 +08:00
parent 9da5526659
commit 6ebff23cb9
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
egs/librispeech/ASR/pruned_transducer_stateless7/conformer.py
View File

@ -450,8 +450,8 @@ class ConformerEncoder(nn.Module):
initial_layerdrop_prob = 0.5
final_layerdrop_prob = 0.05
if warmup_count < 100.0:
# As a special case, if warmup_count < 100.0 return 0 (drop no
if warmup_count < 5.0:
# As a special case, if warmup_count < 5.0 return 0 (drop no
# layers). This is rather ugly, I'm afraid; it is intended to
# enable our scan_pessimistic_batches_for_oom() code to work correctly
# so if we are going to get OOM it will happen early.