diff --git a/README.md b/README.md index 17c49aa30b6f8e1eab777582762ed58716f2de03..1eeca65af9de87f38526b28dad87a24f32b5cf3d 100644 --- a/README.md +++ b/README.md @@ -26,16 +26,16 @@ model-index: metrics: - name: Accuracy type: accuracy - value: 0.9852941176470589 + value: 0.991941391941392 - name: F1 type: f1 - value: 0.9853150765112866 + value: 0.9919569277165429 - name: Precision type: precision - value: 0.9853868369053048 + value: 0.9920048531706146 - name: Recall type: recall - value: 0.9852941176470589 + value: 0.991941391941392 --- This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the audiofolder dataset. It achieves the following results on the evaluation set: -- Loss: 0.0332 -- Accuracy: 0.9853 -- F1: 0.9853 -- Precision: 0.9854 -- Recall: 0.9853 +- Loss: 0.0408 +- Accuracy: 0.9919 +- F1: 0.9920 +- Precision: 0.9920 +- Recall: 0.9919 ## Model description @@ -77,22 +77,19 @@ The following hyperparameters were used during training: - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: cosine - lr_scheduler_warmup_ratio: 0.001 -- num_epochs: 10 +- num_epochs: 7 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:| -| No log | 0.9412 | 12 | 0.1931 | 0.9363 | 0.9365 | 0.9372 | 0.9363 | -| No log | 1.9608 | 25 | 0.0950 | 0.9706 | 0.9704 | 0.9710 | 0.9706 | -| No log | 2.9804 | 38 | 0.0611 | 0.9804 | 0.9804 | 0.9804 | 0.9804 | -| No log | 4.0 | 51 | 0.0492 | 0.9853 | 0.9853 | 0.9853 | 0.9853 | -| No log | 4.9412 | 63 | 0.0588 | 0.9804 | 0.9805 | 0.9814 | 0.9804 | -| No log | 5.9608 | 76 | 0.0368 | 0.9853 | 0.9853 | 0.9854 | 0.9853 | -| No log | 6.9804 | 89 | 0.0382 | 0.9902 | 0.9902 | 0.9903 | 0.9902 | -| No log | 8.0 | 102 | 0.0318 | 0.9951 | 0.9951 | 0.9951 | 0.9951 | -| No log | 8.9412 | 114 | 0.0331 | 0.9853 | 0.9853 | 0.9854 | 0.9853 | -| No log | 9.4118 | 120 | 0.0332 | 0.9853 | 0.9853 | 0.9854 | 0.9853 | +| No log | 0.9956 | 85 | 0.0736 | 0.9788 | 0.9788 | 0.9790 | 0.9788 | +| No log | 1.9912 | 170 | 0.0680 | 0.9758 | 0.9760 | 0.9770 | 0.9758 | +| No log | 2.9985 | 256 | 0.0447 | 0.9875 | 0.9876 | 0.9876 | 0.9875 | +| No log | 3.9941 | 341 | 0.0452 | 0.9905 | 0.9905 | 0.9905 | 0.9905 | +| No log | 4.9898 | 426 | 0.0439 | 0.9919 | 0.9920 | 0.9920 | 0.9919 | +| 0.053 | 5.9971 | 512 | 0.0401 | 0.9919 | 0.9920 | 0.9920 | 0.9919 | +| 0.053 | 6.9693 | 595 | 0.0408 | 0.9919 | 0.9920 | 0.9920 | 0.9919 | ### Framework versions diff --git a/checkpoint-170/config.json b/checkpoint-170/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-170/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-170/model.safetensors b/checkpoint-170/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..77506a0974c618fb6e9ef9883076f7de406a2062 --- /dev/null +++ b/checkpoint-170/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14ad2529c107706df5e38d4e839c5fe54bbd3638840f1b016ba56f9e40cfd45a +size 94763496 diff --git a/checkpoint-170/optimizer.pt b/checkpoint-170/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..7291d24bff80899d3ae094a6d9ce9bbe38bfe74a --- /dev/null +++ b/checkpoint-170/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:feca43a53b1ea22e785cf6aebc67731d4d759491623447479b488904538b9c5d +size 189552570 diff --git a/checkpoint-170/preprocessor_config.json b/checkpoint-170/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-170/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-170/rng_state.pth b/checkpoint-170/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..1f8057471c650bde3cc53ebec371d10395b5dc81 --- /dev/null +++ b/checkpoint-170/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:309f3425e8cb713e30160da01fdcf088ad760dfa7761c7bd73a907a4115b0ee0 +size 14308 diff --git a/checkpoint-170/scheduler.pt b/checkpoint-170/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4862f583aaed37ec1a33b1945d79d5fb64c92df0 --- /dev/null +++ b/checkpoint-170/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e9632dd1819b93de48e3bbe467fe0d96e895179dd60318df68d501d078c6ed9 +size 1064 diff --git a/checkpoint-170/trainer_state.json b/checkpoint-170/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..960285647bd9ace87bc5b55d7b6b81a92fad3bb1 --- /dev/null +++ b/checkpoint-170/trainer_state.json @@ -0,0 +1,66 @@ +{ + "best_metric": 0.9787545787545787, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-85", + "epoch": 1.9912152269399708, + "eval_steps": 500, + "global_step": 170, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9760458977669253, + "eval_loss": 0.0680176243185997, + "eval_precision": 0.9770364666252971, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9805, + "eval_samples_per_second": 342.924, + "eval_steps_per_second": 42.96, + "step": 170 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2.483401519872e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-170/training_args.bin b/checkpoint-170/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-170/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/checkpoint-255/config.json b/checkpoint-255/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-255/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-255/model.safetensors b/checkpoint-255/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e168c3cd0a55a800d6673116822cc65762eb6889 --- /dev/null +++ b/checkpoint-255/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f4d1bcece15f285a46eb855f203bd43924522bb36fe48d2d6c26990580d046d +size 94763496 diff --git a/checkpoint-255/optimizer.pt b/checkpoint-255/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..dec3574839ddbb6d0f4949c0b00dc064fa9f22f3 --- /dev/null +++ b/checkpoint-255/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:55c75ae0d01ca9c0fa93dadeceec95f962b2525037b69dc1a327f1b6c3c0ad59 +size 189552570 diff --git a/checkpoint-255/preprocessor_config.json b/checkpoint-255/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-255/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-255/rng_state.pth b/checkpoint-255/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..f355244972e6058d01893db631d7c135a2372ebd --- /dev/null +++ b/checkpoint-255/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32bb7f9c3cdfdb4246f58a8319fd0b86a841985e3d321ecf44d228390d31634d +size 14308 diff --git a/checkpoint-255/scheduler.pt b/checkpoint-255/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..40a5af11361adaf987f8533da17fb16bb1bac2eb --- /dev/null +++ b/checkpoint-255/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7859ab68e5739b704ca9ce13610ef11b9b1125824d2905d10d4a9ea15d8739ea +size 1064 diff --git a/checkpoint-255/trainer_state.json b/checkpoint-255/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..77cb41221bc6ceab17b53aee7c8ea365519d516d --- /dev/null +++ b/checkpoint-255/trainer_state.json @@ -0,0 +1,78 @@ +{ + "best_metric": 0.989010989010989, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-255", + "epoch": 2.986822840409956, + "eval_steps": 500, + "global_step": 255, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9759605689530879, + "eval_loss": 0.07685838639736176, + "eval_precision": 0.9763775351444481, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9273, + "eval_samples_per_second": 347.571, + "eval_steps_per_second": 43.542, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9875457875457876, + "eval_f1": 0.9875697973801115, + "eval_loss": 0.04443557932972908, + "eval_precision": 0.9876259880187351, + "eval_recall": 0.9875457875457876, + "eval_runtime": 3.9441, + "eval_samples_per_second": 346.088, + "eval_steps_per_second": 43.356, + "step": 170 + }, + { + "epoch": 2.986822840409956, + "eval_accuracy": 0.989010989010989, + "eval_f1": 0.9890405015532383, + "eval_loss": 0.03978995233774185, + "eval_precision": 0.9891330367917903, + "eval_recall": 0.989010989010989, + "eval_runtime": 3.9868, + "eval_samples_per_second": 342.378, + "eval_steps_per_second": 42.891, + "step": 255 + } + ], + "logging_steps": 500, + "max_steps": 255, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 3.7096378747392e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-255/training_args.bin b/checkpoint-255/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..e61b12c0e23d90f0c2542b9fa1cb2858ffc79856 --- /dev/null +++ b/checkpoint-255/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37edd9986d3659d5cfeb9aa8cfbe6429149a6cd74482bcfbbbae5877c8eae7df +size 5240 diff --git a/checkpoint-256/config.json b/checkpoint-256/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-256/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-256/model.safetensors b/checkpoint-256/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ebc43192ea1c14162bbf2e7c7c9b7323be2a3e7e --- /dev/null +++ b/checkpoint-256/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a88f787681237fee33b288f12593aacb934387bd6da2bb8ff790e9b5a05f9556 +size 94763496 diff --git a/checkpoint-256/optimizer.pt b/checkpoint-256/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..296bc2c6a72a864601dd6b0f98b228aaab72c6fc --- /dev/null +++ b/checkpoint-256/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:346b3d592afd4162ca6d18904ccc574e4d170d2fe28c5c444a687f9850fd23b6 +size 189552570 diff --git a/checkpoint-256/preprocessor_config.json b/checkpoint-256/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-256/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-256/rng_state.pth b/checkpoint-256/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..f0af7a3d19d5ad698877769d85059e48940efd32 --- /dev/null +++ b/checkpoint-256/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:681b1a6a4f0a991d47831ff66346aef2bdf16d4f5cc0b1dfbedc7d47e4c328a1 +size 14308 diff --git a/checkpoint-256/scheduler.pt b/checkpoint-256/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..234bfdae5bd9b692b4ee9c9695423adec1c4d8f2 --- /dev/null +++ b/checkpoint-256/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93dfa4ba2e0507f7a6d68626387a161a1bbe6b94f9a9daf967637af5f5748293 +size 1064 diff --git a/checkpoint-256/trainer_state.json b/checkpoint-256/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..517be3385424d1c39a03109232bd0b296a633c48 --- /dev/null +++ b/checkpoint-256/trainer_state.json @@ -0,0 +1,78 @@ +{ + "best_metric": 0.9875457875457876, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-256", + "epoch": 2.998535871156662, + "eval_steps": 500, + "global_step": 256, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9760458977669253, + "eval_loss": 0.0680176243185997, + "eval_precision": 0.9770364666252971, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9805, + "eval_samples_per_second": 342.924, + "eval_steps_per_second": 42.96, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9875457875457876, + "eval_f1": 0.987550637300988, + "eval_loss": 0.044689420610666275, + "eval_precision": 0.9875567820840008, + "eval_recall": 0.9875457875457876, + "eval_runtime": 4.1011, + "eval_samples_per_second": 332.839, + "eval_steps_per_second": 41.696, + "step": 256 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.725102279808e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-256/training_args.bin b/checkpoint-256/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-256/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/checkpoint-341/config.json b/checkpoint-341/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-341/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-341/model.safetensors b/checkpoint-341/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..09179f3e4f9bd17ebfc11dde426ed7437a201659 --- /dev/null +++ b/checkpoint-341/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d274d762a507f6e75c1d2a58abb70f2a03423ada1370bde55f03f89babb022da +size 94763496 diff --git a/checkpoint-341/optimizer.pt b/checkpoint-341/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..675b0a22df1d3579b184a03b178e90634d2a1df2 --- /dev/null +++ b/checkpoint-341/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab5a9612889baf3497db6846ceb8c24a7fa052bc1a8f788d7b5b366a70c8d5b9 +size 189552570 diff --git a/checkpoint-341/preprocessor_config.json b/checkpoint-341/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-341/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-341/rng_state.pth b/checkpoint-341/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..7259f61ec8daf9726ab8636e5454af58c41acf34 --- /dev/null +++ b/checkpoint-341/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c14b301b5ea3ffdd05cbe8ff85a0d16fa4c4467786db8506b5bdcb5af8a9a2d +size 14308 diff --git a/checkpoint-341/scheduler.pt b/checkpoint-341/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..efe43cde99f4286ac87b22b929aaa4301af4698c --- /dev/null +++ b/checkpoint-341/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db61ebcfff67389442b8098d4cae839f684a87068fe21401504beb7ddd27d04d +size 1064 diff --git a/checkpoint-341/trainer_state.json b/checkpoint-341/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..5b90ac919c40038d0dba97da00abb3ddd2dfdae7 --- /dev/null +++ b/checkpoint-341/trainer_state.json @@ -0,0 +1,90 @@ +{ + "best_metric": 0.9904761904761905, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-341", + "epoch": 3.994143484626647, + "eval_steps": 500, + "global_step": 341, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9760458977669253, + "eval_loss": 0.0680176243185997, + "eval_precision": 0.9770364666252971, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9805, + "eval_samples_per_second": 342.924, + "eval_steps_per_second": 42.96, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9875457875457876, + "eval_f1": 0.987550637300988, + "eval_loss": 0.044689420610666275, + "eval_precision": 0.9875567820840008, + "eval_recall": 0.9875457875457876, + "eval_runtime": 4.1011, + "eval_samples_per_second": 332.839, + "eval_steps_per_second": 41.696, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9904761904761905, + "eval_f1": 0.9904945509377323, + "eval_loss": 0.045228052884340286, + "eval_precision": 0.9905452314533213, + "eval_recall": 0.9904761904761905, + "eval_runtime": 3.9393, + "eval_samples_per_second": 346.509, + "eval_steps_per_second": 43.409, + "step": 341 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.966803039744e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-341/training_args.bin b/checkpoint-341/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-341/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/checkpoint-426/config.json b/checkpoint-426/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-426/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-426/model.safetensors b/checkpoint-426/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b47cf29d4974efcc2d310aa700ba4749b22632c6 --- /dev/null +++ b/checkpoint-426/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94760f7df9dc2c6bad702593ec0871720fa6ae352e88f79c71bcfccc8fab1a09 +size 94763496 diff --git a/checkpoint-426/optimizer.pt b/checkpoint-426/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..e8b07ed5da314b2a26319d3fb657a3471ada7468 --- /dev/null +++ b/checkpoint-426/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6550e976738f22f913650571284f8eb12dea30c8877134bbe84a1ebd7a73b5e +size 189552570 diff --git a/checkpoint-426/preprocessor_config.json b/checkpoint-426/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-426/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-426/rng_state.pth b/checkpoint-426/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..b0634373bff4d5562f33f83b1c01ab9107f78c68 --- /dev/null +++ b/checkpoint-426/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2a1aad7f62f9d383e6bd694b87da317b54d942be2b578a455bcdaaf3258d46a +size 14308 diff --git a/checkpoint-426/scheduler.pt b/checkpoint-426/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..5848057b7abfc3bb9ee3960b26437b6b6f868d53 --- /dev/null +++ b/checkpoint-426/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1894ac3a34350cc065802675cae4d32438918a7d772edc033390240aaf5d8841 +size 1064 diff --git a/checkpoint-426/trainer_state.json b/checkpoint-426/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..c4a5bf78e085f4db18734c660539b2fb3cd0231c --- /dev/null +++ b/checkpoint-426/trainer_state.json @@ -0,0 +1,102 @@ +{ + "best_metric": 0.991941391941392, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-426", + "epoch": 4.989751098096632, + "eval_steps": 500, + "global_step": 426, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9760458977669253, + "eval_loss": 0.0680176243185997, + "eval_precision": 0.9770364666252971, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9805, + "eval_samples_per_second": 342.924, + "eval_steps_per_second": 42.96, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9875457875457876, + "eval_f1": 0.987550637300988, + "eval_loss": 0.044689420610666275, + "eval_precision": 0.9875567820840008, + "eval_recall": 0.9875457875457876, + "eval_runtime": 4.1011, + "eval_samples_per_second": 332.839, + "eval_steps_per_second": 41.696, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9904761904761905, + "eval_f1": 0.9904945509377323, + "eval_loss": 0.045228052884340286, + "eval_precision": 0.9905452314533213, + "eval_recall": 0.9904761904761905, + "eval_runtime": 3.9393, + "eval_samples_per_second": 346.509, + "eval_steps_per_second": 43.409, + "step": 341 + }, + { + "epoch": 4.989751098096632, + "eval_accuracy": 0.991941391941392, + "eval_f1": 0.9919569277165429, + "eval_loss": 0.04387320205569267, + "eval_precision": 0.9920048531706146, + "eval_recall": 0.991941391941392, + "eval_runtime": 4.0115, + "eval_samples_per_second": 340.268, + "eval_steps_per_second": 42.627, + "step": 426 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.20850379968e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-426/training_args.bin b/checkpoint-426/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-426/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/checkpoint-512/config.json b/checkpoint-512/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-512/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-512/model.safetensors b/checkpoint-512/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c55c352473b4cc1a0d0d29c34da7de2547a9780d --- /dev/null +++ b/checkpoint-512/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6beddb5fef4d20200cfcf763c9460267deb7ad4514b1fa760fa4f89d398502cf +size 94763496 diff --git a/checkpoint-512/optimizer.pt b/checkpoint-512/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..1ed40a1fc943647adf5af6cd6c38af7274bb7e5f --- /dev/null +++ b/checkpoint-512/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2cfb7299eb13d9980cb44dcbfcf90cc3e4a95d2c84fd840b959005ca0b7566d6 +size 189552570 diff --git a/checkpoint-512/preprocessor_config.json b/checkpoint-512/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-512/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-512/rng_state.pth b/checkpoint-512/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..947d0d41cba5d7eed60e23d174144c754607ae05 --- /dev/null +++ b/checkpoint-512/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:863dd8c5b813736d29c00a5bac8fd9a20f38b88a7fb576cceea0cc85f0cbd789 +size 14308 diff --git a/checkpoint-512/scheduler.pt b/checkpoint-512/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..210d846be25edddf75a04893d20df8827db68826 --- /dev/null +++ b/checkpoint-512/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7900f26d6583fa63e5c081f39e0457bc3033e9e087a3f2085f881f9fac700fb3 +size 1064 diff --git a/checkpoint-512/trainer_state.json b/checkpoint-512/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..e9ac03a7719fa611c37c129c2edb4298bcc35c1b --- /dev/null +++ b/checkpoint-512/trainer_state.json @@ -0,0 +1,121 @@ +{ + "best_metric": 0.991941391941392, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-426", + "epoch": 5.997071742313324, + "eval_steps": 500, + "global_step": 512, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9760458977669253, + "eval_loss": 0.0680176243185997, + "eval_precision": 0.9770364666252971, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9805, + "eval_samples_per_second": 342.924, + "eval_steps_per_second": 42.96, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9875457875457876, + "eval_f1": 0.987550637300988, + "eval_loss": 0.044689420610666275, + "eval_precision": 0.9875567820840008, + "eval_recall": 0.9875457875457876, + "eval_runtime": 4.1011, + "eval_samples_per_second": 332.839, + "eval_steps_per_second": 41.696, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9904761904761905, + "eval_f1": 0.9904945509377323, + "eval_loss": 0.045228052884340286, + "eval_precision": 0.9905452314533213, + "eval_recall": 0.9904761904761905, + "eval_runtime": 3.9393, + "eval_samples_per_second": 346.509, + "eval_steps_per_second": 43.409, + "step": 341 + }, + { + "epoch": 4.989751098096632, + "eval_accuracy": 0.991941391941392, + "eval_f1": 0.9919569277165429, + "eval_loss": 0.04387320205569267, + "eval_precision": 0.9920048531706146, + "eval_recall": 0.991941391941392, + "eval_runtime": 4.0115, + "eval_samples_per_second": 340.268, + "eval_steps_per_second": 42.627, + "step": 426 + }, + { + "epoch": 5.856515373352855, + "grad_norm": 0.01621050015091896, + "learning_rate": 6.1795711069424666e-06, + "loss": 0.053, + "step": 500 + }, + { + "epoch": 5.997071742313324, + "eval_accuracy": 0.991941391941392, + "eval_f1": 0.9919507596926951, + "eval_loss": 0.04010434448719025, + "eval_precision": 0.991971823048746, + "eval_recall": 0.991941391941392, + "eval_runtime": 4.1768, + "eval_samples_per_second": 326.803, + "eval_steps_per_second": 40.94, + "step": 512 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 7.450204559616e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-512/training_args.bin b/checkpoint-512/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-512/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/checkpoint-595/config.json b/checkpoint-595/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-595/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-595/model.safetensors b/checkpoint-595/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..72b66549669408bd2a4da36123818d6aa603f3b3 --- /dev/null +++ b/checkpoint-595/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:820f93ff9f0b4aabd94c37f6b8f8cf46bc811ac597f1682a787fbfde25a51a38 +size 94763496 diff --git a/checkpoint-595/optimizer.pt b/checkpoint-595/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..61e743f7f288f3341492ec2592c97fe6f8527fef --- /dev/null +++ b/checkpoint-595/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5969c4f905799778a874dc0915fa5294dab43984b00fe3f2779de557361db5b +size 189552570 diff --git a/checkpoint-595/preprocessor_config.json b/checkpoint-595/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-595/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-595/rng_state.pth b/checkpoint-595/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..3b5d06e7d3906e5139bb5220b011e5627cf794de --- /dev/null +++ b/checkpoint-595/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aab42b192f76daeb08650fbae0bdaa3c3c9de65636156ae96835a00bcac7a585 +size 14308 diff --git a/checkpoint-595/scheduler.pt b/checkpoint-595/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..d2595586b77e20b6f4aad8c330be0183c485e151 --- /dev/null +++ b/checkpoint-595/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7e9e8d4a532362dab3367ad2982f480a24bfd56b7a4400f3cf522e250890c25 +size 1064 diff --git a/checkpoint-595/trainer_state.json b/checkpoint-595/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..54de2bbd00331cf16ad8f5cdedec08436815d7c0 --- /dev/null +++ b/checkpoint-595/trainer_state.json @@ -0,0 +1,133 @@ +{ + "best_metric": 0.991941391941392, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-426", + "epoch": 6.9692532942898975, + "eval_steps": 500, + "global_step": 595, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9758241758241758, + "eval_f1": 0.9760458977669253, + "eval_loss": 0.0680176243185997, + "eval_precision": 0.9770364666252971, + "eval_recall": 0.9758241758241758, + "eval_runtime": 3.9805, + "eval_samples_per_second": 342.924, + "eval_steps_per_second": 42.96, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9875457875457876, + "eval_f1": 0.987550637300988, + "eval_loss": 0.044689420610666275, + "eval_precision": 0.9875567820840008, + "eval_recall": 0.9875457875457876, + "eval_runtime": 4.1011, + "eval_samples_per_second": 332.839, + "eval_steps_per_second": 41.696, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9904761904761905, + "eval_f1": 0.9904945509377323, + "eval_loss": 0.045228052884340286, + "eval_precision": 0.9905452314533213, + "eval_recall": 0.9904761904761905, + "eval_runtime": 3.9393, + "eval_samples_per_second": 346.509, + "eval_steps_per_second": 43.409, + "step": 341 + }, + { + "epoch": 4.989751098096632, + "eval_accuracy": 0.991941391941392, + "eval_f1": 0.9919569277165429, + "eval_loss": 0.04387320205569267, + "eval_precision": 0.9920048531706146, + "eval_recall": 0.991941391941392, + "eval_runtime": 4.0115, + "eval_samples_per_second": 340.268, + "eval_steps_per_second": 42.627, + "step": 426 + }, + { + "epoch": 5.856515373352855, + "grad_norm": 0.01621050015091896, + "learning_rate": 6.1795711069424666e-06, + "loss": 0.053, + "step": 500 + }, + { + "epoch": 5.997071742313324, + "eval_accuracy": 0.991941391941392, + "eval_f1": 0.9919507596926951, + "eval_loss": 0.04010434448719025, + "eval_precision": 0.991971823048746, + "eval_recall": 0.991941391941392, + "eval_runtime": 4.1768, + "eval_samples_per_second": 326.803, + "eval_steps_per_second": 40.94, + "step": 512 + }, + { + "epoch": 6.9692532942898975, + "eval_accuracy": 0.991941391941392, + "eval_f1": 0.9919569277165429, + "eval_loss": 0.040794000029563904, + "eval_precision": 0.9920048531706146, + "eval_recall": 0.991941391941392, + "eval_runtime": 3.9134, + "eval_samples_per_second": 348.802, + "eval_steps_per_second": 43.696, + "step": 595 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 8.6546088132096e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-595/training_args.bin b/checkpoint-595/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-595/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/checkpoint-85/config.json b/checkpoint-85/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-85/config.json @@ -0,0 +1,80 @@ +{ + "_name_or_path": "ntu-spml/distilhubert", + "activation_dropout": 0.1, + "apply_spec_augment": false, + "architectures": [ + "HubertForSequenceClassification" + ], + "attention_dropout": 0.1, + "bos_token_id": 1, + "classifier_proj_size": 256, + "conv_bias": false, + "conv_dim": [ + 512, + 512, + 512, + 512, + 512, + 512, + 512 + ], + "conv_kernel": [ + 10, + 3, + 3, + 3, + 3, + 2, + 2 + ], + "conv_stride": [ + 5, + 2, + 2, + 2, + 2, + 2, + 2 + ], + "ctc_loss_reduction": "sum", + "ctc_zero_infinity": false, + "do_stable_layer_norm": false, + "eos_token_id": 2, + "feat_extract_activation": "gelu", + "feat_extract_norm": "group", + "feat_proj_dropout": 0.0, + "feat_proj_layer_norm": false, + "final_dropout": 0.0, + "hidden_act": "gelu", + "hidden_dropout": 0.1, + "hidden_size": 768, + "id2label": { + "0": "crying", + "1": "no_crying" + }, + "initializer_range": 0.02, + "intermediate_size": 3072, + "label2id": { + "crying": "0", + "no_crying": "1" + }, + "layer_norm_eps": 1e-05, + "layerdrop": 0.0, + "mask_feature_length": 10, + "mask_feature_min_masks": 0, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_masks": 2, + "mask_time_prob": 0.05, + "model_type": "hubert", + "num_attention_heads": 12, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 2, + "pad_token_id": 0, + "torch_dtype": "float32", + "transformers_version": "4.44.2", + "use_weighted_layer_sum": false, + "vocab_size": 32 +} diff --git a/checkpoint-85/model.safetensors b/checkpoint-85/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a4a245229c8de700b413ffe10c8d03158e50ed5d --- /dev/null +++ b/checkpoint-85/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:687e267a3c37d18cce5591d5301b06a4193f540a3ed0c054e9913e5e610aef05 +size 94763496 diff --git a/checkpoint-85/optimizer.pt b/checkpoint-85/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..e980b6c8ea2b4bc57a62f661e0cf03de0369051b --- /dev/null +++ b/checkpoint-85/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ff974afccdd073da4ec78806e9470c99c469ae8a871771439016949486039a9 +size 189552570 diff --git a/checkpoint-85/preprocessor_config.json b/checkpoint-85/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ad80e2bc82023595cde1d13979733ba2dca60cc --- /dev/null +++ b/checkpoint-85/preprocessor_config.json @@ -0,0 +1,9 @@ +{ + "do_normalize": false, + "feature_extractor_type": "Wav2Vec2FeatureExtractor", + "feature_size": 1, + "padding_side": "right", + "padding_value": 0, + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-85/rng_state.pth b/checkpoint-85/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..4c5ac8c3689001df6eecfdf0e94709303407ad2a --- /dev/null +++ b/checkpoint-85/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc2470f568405de0c2e468ca2da81de2e52f44f37dfaf8836f660441f648fa7a +size 14308 diff --git a/checkpoint-85/scheduler.pt b/checkpoint-85/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..fbad7a97a88117443a8a9d272b29a6c604cb6f57 --- /dev/null +++ b/checkpoint-85/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:daf90af7d8b581c8b4882d01681a4f4fb886f79b4c5624a5d256aba635f56702 +size 1064 diff --git a/checkpoint-85/trainer_state.json b/checkpoint-85/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..c48612280f108c8e33785720e1aab136b737ec2d --- /dev/null +++ b/checkpoint-85/trainer_state.json @@ -0,0 +1,54 @@ +{ + "best_metric": 0.9787545787545787, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-85", + "epoch": 0.9956076134699854, + "eval_steps": 500, + "global_step": 85, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9787545787545787, + "eval_f1": 0.9788275754377449, + "eval_loss": 0.07363971322774887, + "eval_precision": 0.9790028856592324, + "eval_recall": 0.9787545787545787, + "eval_runtime": 4.1234, + "eval_samples_per_second": 331.041, + "eval_steps_per_second": 41.471, + "step": 85 + } + ], + "logging_steps": 500, + "max_steps": 595, + "num_input_tokens_seen": 0, + "num_train_epochs": 7, + "save_steps": 500, + "stateful_callbacks": { + "EarlyStoppingCallback": { + "args": { + "early_stopping_patience": 3, + "early_stopping_threshold": 0.0 + }, + "attributes": { + "early_stopping_patience_counter": 0 + } + }, + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.241700759936e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-85/training_args.bin b/checkpoint-85/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..8862d582702297ee422932b621da79a72d7db40c --- /dev/null +++ b/checkpoint-85/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +size 5240 diff --git a/model.safetensors b/model.safetensors index b63e1b217d7bda6f42a6ac5fe5cc72b618e9bffe..b47cf29d4974efcc2d310aa700ba4749b22632c6 100644 --- a/model.safetensors +++ b/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:098c90d2cba195ac20033090c5954f7ef596abcdeebba724cb1365151033b05e +oid sha256:94760f7df9dc2c6bad702593ec0871720fa6ae352e88f79c71bcfccc8fab1a09 size 94763496 diff --git a/runs/Sep02_00-52-32_ubumarcos/events.out.tfevents.1725231165.ubumarcos b/runs/Sep02_00-52-32_ubumarcos/events.out.tfevents.1725231165.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..395a5b59fb46d5dc3efceef260c2c9bc9667ceec --- /dev/null +++ b/runs/Sep02_00-52-32_ubumarcos/events.out.tfevents.1725231165.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f32bcf30faf64ec38ca7d26a191ea5b4533e431c3825aeee43b9ba13d061e1a9 +size 4182 diff --git a/runs/Sep02_00-55-46_ubumarcos/events.out.tfevents.1725231353.ubumarcos b/runs/Sep02_00-55-46_ubumarcos/events.out.tfevents.1725231353.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..489b0b3a0f084529396cbf0a8c5cee5c32a79268 --- /dev/null +++ b/runs/Sep02_00-55-46_ubumarcos/events.out.tfevents.1725231353.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b767b0f06fb23734f9483dea11ffba8066a8f47465ec550e8af215bc57acdc5c +size 6620 diff --git a/runs/Sep02_01-09-27_ubumarcos/events.out.tfevents.1725232180.ubumarcos b/runs/Sep02_01-09-27_ubumarcos/events.out.tfevents.1725232180.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..10915a738baaab67b78a687a94085f3d80b1fe1a --- /dev/null +++ b/runs/Sep02_01-09-27_ubumarcos/events.out.tfevents.1725232180.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49995861ce24f79defd1239b0a9255f1722c4193a1d606c6c2355783b9200a96 +size 7570 diff --git a/runs/Sep02_11-35-27_ubumarcos/events.out.tfevents.1725269735.ubumarcos b/runs/Sep02_11-35-27_ubumarcos/events.out.tfevents.1725269735.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..bd423e8c1bcd18e92688212eed864db9ed43176c --- /dev/null +++ b/runs/Sep02_11-35-27_ubumarcos/events.out.tfevents.1725269735.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eea5820ea966730d8e65a4c150b9a7ba53beaed1521273e658e1d9f112bad09a +size 9669 diff --git a/training_args.bin b/training_args.bin index ff5c682679da830384645399a104cd532805ce20..8862d582702297ee422932b621da79a72d7db40c 100644 --- a/training_args.bin +++ b/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:3b77b2a33ca88330b7909aacab81332695e9c9dcf797e7a7757ca31ffcde6f31 +oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 size 5240