This repository has been archived by the owner on Sep 19, 2024. It is now read-only.
generated from rochacbruno/python-project-template
-
Notifications
You must be signed in to change notification settings - Fork 6
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
22 changed files
with
6,891 additions
and
181 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,48 +1,9 @@ | ||
| trainer: | ||
| callbacks: | ||
| - class_path: scprint.trainer.TrainingMode | ||
| init_args: | ||
| do_denoise: True | ||
| noise: | ||
| - 0.6 | ||
| do_cce: True | ||
| do_ecs: False | ||
| do_mvc: False | ||
| do_generate: True | ||
| do_adv_cls: False | ||
| do_next_tp: False | ||
| do_adv_batch: False | ||
| run_full_forward: False | ||
| do_cls: True | ||
| class_scale: 1 | ||
| warmup_duration: 500 | ||
| fused_adam: True | ||
| mask_ratio: [] | ||
| - class_path: lightning.pytorch.callbacks.StochasticWeightAveraging | ||
| init_args: | ||
| swa_lrs: 0.03 | ||
| - class_path: lightning.pytorch.callbacks.ModelCheckpoint | ||
| init_args: | ||
| monitor: val_loss | ||
| save_top_k: 6 | ||
| save_last: True | ||
| - class_path: lightning.pytorch.callbacks.EarlyStopping | ||
| init_args: | ||
| monitor: val_loss | ||
| patience: 3 | ||
| - class_path: lightning.pytorch.callbacks.LearningRateMonitor | ||
| init_args: | ||
| logging_interval: epoch | ||
| strategy: ddp_find_unused_parameters_true | ||
| model: | ||
| nhead: 8 | ||
| nlayers: 16 #used to be 12 | ||
| layers_cls: [512] | ||
| d_model: 512 | ||
| freeze_embeddings: False | ||
| data: | ||
| max_len: 2200 | ||
| weight_scaler: 100 | ||
| train_oversampling_per_epoch: 0.5 | ||
| batch_size: 16 | ||
| collection_name: all no zhang13M | ||
| num_workers: 12 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,20 +1,7 @@ | ||
| trainer: | ||
| limit_train_batches: 5000 | ||
| limit_val_batches: 1000 | ||
| max_time: | ||
| hours: 72 | ||
| model: | ||
| nhead: 4 | ||
| nlayers: 8 | ||
| freeze_embeddings: True | ||
| layers_cls: [256] | ||
| d_model: 256 | ||
| data: | ||
| collection_name: all no zhang13M #preprocessed dataset | ||
| max_len: 2200 | ||
| weight_scaler: 40 | ||
| train_oversampling_per_epoch: 0.4 | ||
| validation_split: 0.02 | ||
| test_split: 0.02 | ||
| batch_size: 64 | ||
| num_workers: 12 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,68 +1,17 @@ | ||
| trainer: | ||
| strategy: ddp_find_unused_parameters_true | ||
| num_nodes: 1 | ||
| max_time: | ||
| hours: 72 | ||
| log_every_n_steps: 300 | ||
| precision: 16-mixed | ||
| gradient_clip_val: 500 | ||
| limit_train_batches: 36000 | ||
| limit_val_batches: 4000 | ||
| limit_train_batches: 14000 | ||
| limit_val_batches: 2000 | ||
| reload_dataloaders_every_n_epochs: 1 | ||
| accumulate_grad_batches: 6 | ||
| callbacks: | ||
| - class_path: scprint.trainer.TrainingMode | ||
| init_args: | ||
| do_denoise: True | ||
| noise: | ||
| - 0.6 | ||
| do_cce: True | ||
| do_ecs: False | ||
| do_mvc: False | ||
| do_generate: True | ||
| do_adv_cls: False | ||
| do_next_tp: False | ||
| do_adv_batch: False | ||
| run_full_forward: False | ||
| do_cls: True | ||
| class_scale: 1 | ||
| warmup_duration: 1000 # prev 3000 (maybe it helped?) | ||
| fused_adam: True | ||
| mask_ratio: [] | ||
| #- class_path: lightning.pytorch.callbacks.StochasticWeightAveraging | ||
| # init_args: | ||
| # swa_lrs: 0.0 | ||
| - class_path: lightning.pytorch.callbacks.ModelCheckpoint | ||
| init_args: | ||
| monitor: val_loss | ||
| save_top_k: 6 | ||
| save_last: True | ||
| - class_path: lightning.pytorch.callbacks.EarlyStopping | ||
| init_args: | ||
| monitor: val_loss | ||
| patience: 8 | ||
| - class_path: lightning.pytorch.callbacks.LearningRateMonitor | ||
| init_args: | ||
| logging_interval: epoch | ||
| #- class_path: lightning.pytorch.callbacks.LearningRateFinder | ||
| #init_args: | ||
| # mode: exponential | ||
|
|
||
| #plugins: | ||
| # - class_path: lightning.pytorch.plugins.environments.SLURMEnvironment | ||
| # requeue_signal: signal.SIGHUP | ||
| accumulate_grad_batches: 2 | ||
| model: | ||
| nhead: 20 | ||
| lr: 0.00001 | ||
| nlayers: 32 | ||
| layers_cls: [512] | ||
| d_model: 1280 | ||
| freeze_embeddings: True | ||
| data: | ||
| collection_name: all no zhang13M #preprocessed dataset #all no zhang13M | ||
| how: random expr | ||
| max_len: 2200 | ||
| weight_scaler: 2000 | ||
| train_oversampling_per_epoch: 1 | ||
| batch_size: 3 | ||
| num_workers: 9 |
Oops, something went wrong.