1800 Commits

Author SHA1 Message Date
Daniel Povey
1d40239d69 Merge branch 'scaled_adam_exp872' into scaled_adam_exp873 2023-01-09 14:52:48 +08:00
Daniel Povey
e739d8aa38 Fix layer_skip_rate so it's actually used, increase its value. 2023-01-09 13:34:32 +08:00
Daniel Povey
1a0155fcb5 Merge branch 'scaled_adam_exp863' into scaled_adam_exp870
# Conflicts:
#	egs/librispeech/ASR/pruned_transducer_stateless7/scaling.py
2023-01-08 23:36:29 +08:00
Daniel Povey
326cb75033 Increase layer_skip_rate slightly 2023-01-08 15:48:23 +08:00
Daniel Povey
62b42887b4 Revert zipformer.py to status on previous commit 2023-01-08 13:17:39 +08:00
Daniel Povey
e952598677 Merge branch 'scaled_adam_exp846' into scaled_adam_exp866 2023-01-08 13:16:24 +08:00
Daniel Povey
117db124d0 Implement higher layerdrop for central stacks 2023-01-08 13:16:10 +08:00
Daniel Povey
c7107ead64 Fix bug in get_adjusted_batch_count 2023-01-07 17:45:22 +08:00
Daniel Povey
b3527fe4ac Implement caching evaluation for ConvNeXt 2023-01-07 17:31:20 +08:00
Daniel Povey
9242800d42 Remove the 8x-subsampled stack 2023-01-07 12:59:57 +08:00
Daniel Povey
ef48019d6e Reduce feedforward-dims 2023-01-06 22:26:58 +08:00
Daniel Povey
9b0c0aabb2 Merge branch 'scaled_adam_exp829' into scaled_adam_exp860
# Conflicts:
#	egs/librispeech/ASR/pruned_transducer_stateless7/zipformer.py
2023-01-06 22:24:45 +08:00
Daniel Povey
6a762914bf Increase base-lr from 0.05 t to 0.055 2023-01-06 13:35:57 +08:00
Daniel Povey
5564a0efb0 Further tune lr scales; increase base-lr 2023-01-06 13:34:48 +08:00
Daniel Povey
f6f088489d Adjust lr_scales, make them closer to 1. 2023-01-05 23:49:42 +08:00
Daniel Povey
ccc38a97f7 Reduce lr_scales of soem sub modules 2023-01-05 18:50:04 +08:00
Daniel Povey
90c02b471c Revert base LR to 0.05 2023-01-05 16:27:43 +08:00
Daniel Povey
067b861c70 Use largest LR for printing 2023-01-05 14:46:15 +08:00
Daniel Povey
6c7fd8c046 Increase base-lr to 0.06 2023-01-05 14:23:59 +08:00
Daniel Povey
95e8296014 Use downsampling_factor ** -0.333 as the scale for stacks 2023-01-05 14:23:40 +08:00
Daniel Povey
0d7161ebec Use get_parameter_groups_with_lr in train.py; bug fixes 2023-01-05 14:11:33 +08:00
Daniel Povey
1db509ea31 Attempt to implement slower learning for downsampled modules 2023-01-05 13:39:22 +08:00
Daniel Povey
b7be18c2f8 Keep only needed changes from Liyong's branch 2023-01-05 12:23:32 +08:00
Daniel Povey
096ebeaf23 take a couple files from liyong's branch 2023-01-05 12:01:42 +08:00
Daniel Povey
22b4a417dd Implement extra_layerdrop 2023-01-04 20:59:58 +08:00
Daniel Povey
b973929d7c Bug fixes to ScheduledFloat 2023-01-04 20:54:05 +08:00
Daniel Povey
ae73469b7e Refactor ScheduledFloat to include PiecewiseLinear 2023-01-04 20:46:42 +08:00
Daniel Povey
f688066517 Merge branch 'scaled_adam_exp823' into scaled_adam_exp843 2023-01-04 17:02:37 +08:00
Daniel Povey
f7d67f5456 Higher dropout schedule for SmallConvolutionModule 2023-01-02 14:58:23 +08:00
Daniel Povey
5223286424 Add SmallConvolutionModule 2023-01-02 14:47:28 +08:00
Daniel Povey
a61bd01e5b convnext1 kernel size 5, 5 to 5, 7 2023-01-02 14:17:51 +08:00
Daniel Povey
e4d0ac0946 Let the feedforward dims be respectively 3*feedforward_dim // 4 and 5*feedforward_dim // 4. 2023-01-02 00:24:12 +08:00
Daniel Povey
3a5b3f640d Remove eps from BasicNorm and reintroduce bias 2023-01-02 00:02:31 +08:00
Daniel Povey
a2227a07fc Revert some changes to Balancer. 2023-01-01 23:02:52 +08:00
Daniel Povey
e52bfb7219 Revert final conv_skip_rate from 0.01 to 0.0 2023-01-01 22:13:13 +08:00
Daniel Povey
460fb945ec Merge branch 'scaled_adam_exp813' into scaled_adam_exp820 2023-01-01 22:12:10 +08:00
Daniel Povey
977d412690 Cosmetic fix 2023-01-01 21:43:14 +08:00
Daniel Povey
dadeb3feec Fixes for jit scripting and osmetic improvements 2023-01-01 14:35:51 +08:00
Daniel Povey
60d491eee6 Bug fix 2023-01-01 14:31:28 +08:00
Daniel Povey
287bd120be Reduce min_abs of zipformer balancer1; constraints on eps of Conv2dSubsampling.out_norm 2023-01-01 14:28:18 +08:00
Daniel Povey
1797d0ec6d Fix bugs in how max_rms/min_rms constraint were applied, which had the effect of making min_rms dominate over mean. 2023-01-01 13:05:41 +08:00
Daniel Povey
8db0636f1d Fix to Balancer to treat max-rms and min-rms losses separately, only max-rms loss scaled up 2023-01-01 00:38:07 +08:00
Daniel Povey
907d28ca2a Make RMS loss dominate mean loss in Balancer if both are active; remove the 4x scale introduced in 814. 2023-01-01 00:09:14 +08:00
Daniel Povey
a2815ea0df Increase max_abs of ZipformerEncoderLayer.balancer2 from 1.0 to 4.0. 2023-01-01 00:00:26 +08:00
Daniel Povey
63472a19b1 Whitespace fix 2022-12-31 23:50:09 +08:00
Daniel Povey
008dbaf745 Use 4 times the normal grad_scale for BasicNorm if max_rms violated. 2022-12-31 23:42:38 +08:00
Daniel Povey
4a4d12c994 Revert kernel size of convnext2 from 5x5 to 7x7 2022-12-31 21:52:11 +08:00
Daniel Povey
d0ae60400e Decrease convnext1 kernel size from 7x7 to 5x5 2022-12-31 17:19:02 +08:00
Daniel Povey
d48b2ccb45 Reduce kernel size of convnext2 from 7 to 5. 2022-12-31 17:10:31 +08:00
Daniel Povey
c533c30442 Increase final conv_skip_rate from 0.0 to 0.01 2022-12-31 15:10:52 +08:00