diff --git a/README.md b/README.md index 1eeca65af9de87f38526b28dad87a24f32b5cf3d..f32cbc9b53dcf5bd43774994bf7d42c0f392e101 100644 --- a/README.md +++ b/README.md @@ -8,9 +8,6 @@ datasets: - audiofolder metrics: - accuracy -- f1 -- precision -- recall model-index: - name: distilhubert-finetuned-cry-detector results: @@ -21,21 +18,12 @@ model-index: name: audiofolder type: audiofolder config: default - split: train + split: None args: default metrics: - name: Accuracy type: accuracy - value: 0.991941391941392 - - name: F1 - type: f1 - value: 0.9919569277165429 - - name: Precision - type: precision - value: 0.9920048531706146 - - name: Recall - type: recall - value: 0.991941391941392 + value: 0.9926739926739927 --- 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: 0.0408 -- Accuracy: 0.9919 -- F1: 0.9920 -- Precision: 0.9920 -- Recall: 0.9919 +- Loss: 0.0459 +- Accuracy: 0.9927 ## Model description @@ -77,19 +62,21 @@ 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: 7 +- num_epochs: 10 ### Training results -| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | -|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:| -| 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 | +| Training Loss | Epoch | Step | Validation Loss | Accuracy | +|:-------------:|:------:|:----:|:---------------:|:--------:| +| No log | 0.9956 | 85 | 0.0692 | 0.9773 | +| No log | 1.9912 | 170 | 0.0466 | 0.9861 | +| No log | 2.9985 | 256 | 0.0489 | 0.9853 | +| No log | 3.9941 | 341 | 0.0423 | 0.9897 | +| No log | 4.9898 | 426 | 0.0443 | 0.9919 | +| 0.055 | 5.9971 | 512 | 0.0434 | 0.9927 | +| 0.055 | 6.9927 | 597 | 0.0440 | 0.9927 | +| 0.055 | 8.0 | 683 | 0.0460 | 0.9927 | +| 0.055 | 8.9956 | 768 | 0.0459 | 0.9927 | ### Framework versions diff --git a/checkpoint-170/model.safetensors b/checkpoint-170/model.safetensors index 77506a0974c618fb6e9ef9883076f7de406a2062..4dda87edb0e7ca2086b861aa309031d6ab7720c5 100644 --- a/checkpoint-170/model.safetensors +++ b/checkpoint-170/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:14ad2529c107706df5e38d4e839c5fe54bbd3638840f1b016ba56f9e40cfd45a +oid sha256:234766c4455d275002665fd833b44efa39e41bc2661c4bb2462491e9b235265b size 94763496 diff --git a/checkpoint-170/optimizer.pt b/checkpoint-170/optimizer.pt index 7291d24bff80899d3ae094a6d9ce9bbe38bfe74a..05020f0ccc3bef504995b3ff62ffe5b2ef42f213 100644 --- a/checkpoint-170/optimizer.pt +++ b/checkpoint-170/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:feca43a53b1ea22e785cf6aebc67731d4d759491623447479b488904538b9c5d +oid sha256:54168c775d69b037d7dd954b39f148a62b5d5230761ef39bcb3fc60c2c2b65be size 189552570 diff --git a/checkpoint-170/scheduler.pt b/checkpoint-170/scheduler.pt index 4862f583aaed37ec1a33b1945d79d5fb64c92df0..8586668d4d3aa9bbd15cdb7b42394d4c7997054b 100644 --- a/checkpoint-170/scheduler.pt +++ b/checkpoint-170/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:9e9632dd1819b93de48e3bbe467fe0d96e895179dd60318df68d501d078c6ed9 +oid sha256:e2034409269304d3556c2ec17a8e8112c21629214e514ea105efbb26052b23cf size 1064 diff --git a/checkpoint-170/trainer_state.json b/checkpoint-170/trainer_state.json index 960285647bd9ace87bc5b55d7b6b81a92fad3bb1..0a0fccf70076948c4d17bfb499c3bc643fa062ca 100644 --- a/checkpoint-170/trainer_state.json +++ b/checkpoint-170/trainer_state.json @@ -1,6 +1,6 @@ { - "best_metric": 0.9787545787545787, - "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-85", + "best_metric": 0.9860805860805861, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-170", "epoch": 1.9912152269399708, "eval_steps": 500, "global_step": 170, @@ -10,33 +10,27 @@ "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, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, "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, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, "step": 170 } ], "logging_steps": 500, - "max_steps": 595, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 7, + "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { diff --git a/checkpoint-170/training_args.bin b/checkpoint-170/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/checkpoint-170/training_args.bin +++ b/checkpoint-170/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240 diff --git a/checkpoint-256/model.safetensors b/checkpoint-256/model.safetensors index ebc43192ea1c14162bbf2e7c7c9b7323be2a3e7e..0355d2d714e77d35759dbf462bb4644229635b6c 100644 --- a/checkpoint-256/model.safetensors +++ b/checkpoint-256/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a88f787681237fee33b288f12593aacb934387bd6da2bb8ff790e9b5a05f9556 +oid sha256:74664b03f722ebd6f388e4906fa982d1ec05522a8b62f38b65182ae0f78dd659 size 94763496 diff --git a/checkpoint-256/optimizer.pt b/checkpoint-256/optimizer.pt index 296bc2c6a72a864601dd6b0f98b228aaab72c6fc..bf230ea74fcbc155e80cad9f60d5fe4484050430 100644 --- a/checkpoint-256/optimizer.pt +++ b/checkpoint-256/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:346b3d592afd4162ca6d18904ccc574e4d170d2fe28c5c444a687f9850fd23b6 +oid sha256:c93c7b3a43b80636a934d1782de63f8303fd44e61f41d41d41caa10a9639ffc1 size 189552570 diff --git a/checkpoint-256/scheduler.pt b/checkpoint-256/scheduler.pt index 234bfdae5bd9b692b4ee9c9695423adec1c4d8f2..9bf65f88975f1f8f262c36f2f436a418f294d554 100644 --- a/checkpoint-256/scheduler.pt +++ b/checkpoint-256/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:93dfa4ba2e0507f7a6d68626387a161a1bbe6b94f9a9daf967637af5f5748293 +oid sha256:c1daa6dfbc632505773869a612096b9f14303a596de4d72e49c6d934621e4204 size 1064 diff --git a/checkpoint-256/trainer_state.json b/checkpoint-256/trainer_state.json index 517be3385424d1c39a03109232bd0b296a633c48..28dd9e071356d54db7de7083bc5d373f3f291732 100644 --- a/checkpoint-256/trainer_state.json +++ b/checkpoint-256/trainer_state.json @@ -1,6 +1,6 @@ { - "best_metric": 0.9875457875457876, - "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-256", + "best_metric": 0.9860805860805861, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-170", "epoch": 2.998535871156662, "eval_steps": 500, "global_step": 256, @@ -10,45 +10,36 @@ "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, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, "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, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, "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, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, "step": 256 } ], "logging_steps": 500, - "max_steps": 595, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 7, + "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { diff --git a/checkpoint-256/training_args.bin b/checkpoint-256/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/checkpoint-256/training_args.bin +++ b/checkpoint-256/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240 diff --git a/checkpoint-341/model.safetensors b/checkpoint-341/model.safetensors index 09179f3e4f9bd17ebfc11dde426ed7437a201659..bfc0e5bd107954cb2a0aedced68af33505cff094 100644 --- a/checkpoint-341/model.safetensors +++ b/checkpoint-341/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d274d762a507f6e75c1d2a58abb70f2a03423ada1370bde55f03f89babb022da +oid sha256:4c4695016088987cb75a8381886d078dee5da33d439dfa3472cbded298196975 size 94763496 diff --git a/checkpoint-341/optimizer.pt b/checkpoint-341/optimizer.pt index 675b0a22df1d3579b184a03b178e90634d2a1df2..0f0552d9a6a696ef63eed17d5d6cc698788b7a6c 100644 --- a/checkpoint-341/optimizer.pt +++ b/checkpoint-341/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:ab5a9612889baf3497db6846ceb8c24a7fa052bc1a8f788d7b5b366a70c8d5b9 +oid sha256:88625b2c249303af96915509b84fbe0fd810785b84d32d68e936f83e44e0c710 size 189552570 diff --git a/checkpoint-341/scheduler.pt b/checkpoint-341/scheduler.pt index efe43cde99f4286ac87b22b929aaa4301af4698c..906be069a04584b0206917bdc3bda9acbe78311f 100644 --- a/checkpoint-341/scheduler.pt +++ b/checkpoint-341/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:db61ebcfff67389442b8098d4cae839f684a87068fe21401504beb7ddd27d04d +oid sha256:0d0e5f3394e1cb08f5e6ec9d7e11dc43d76ef8c66bf2eaa7a0f848f815b1d4cc size 1064 diff --git a/checkpoint-341/trainer_state.json b/checkpoint-341/trainer_state.json index 5b90ac919c40038d0dba97da00abb3ddd2dfdae7..d68e66fd4394b8f3853d23a78d7b8d0800201584 100644 --- a/checkpoint-341/trainer_state.json +++ b/checkpoint-341/trainer_state.json @@ -1,5 +1,5 @@ { - "best_metric": 0.9904761904761905, + "best_metric": 0.9897435897435898, "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-341", "epoch": 3.994143484626647, "eval_steps": 500, @@ -10,57 +10,45 @@ "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, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, "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, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, "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, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, "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, + "eval_accuracy": 0.9897435897435898, + "eval_loss": 0.04228556528687477, + "eval_runtime": 15.2931, + "eval_samples_per_second": 89.256, + "eval_steps_per_second": 11.181, "step": 341 } ], "logging_steps": 500, - "max_steps": 595, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 7, + "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { diff --git a/checkpoint-341/training_args.bin b/checkpoint-341/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/checkpoint-341/training_args.bin +++ b/checkpoint-341/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240 diff --git a/checkpoint-426/model.safetensors b/checkpoint-426/model.safetensors index b47cf29d4974efcc2d310aa700ba4749b22632c6..17ed7020b5b38c3749565bca9cf64e211c3f97e0 100644 --- a/checkpoint-426/model.safetensors +++ b/checkpoint-426/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:94760f7df9dc2c6bad702593ec0871720fa6ae352e88f79c71bcfccc8fab1a09 +oid sha256:d14ac9c4fea572b7ef6e0e3502f94a78d1c0c3fef295cf4a6f6746aeac8feaf9 size 94763496 diff --git a/checkpoint-426/optimizer.pt b/checkpoint-426/optimizer.pt index e8b07ed5da314b2a26319d3fb657a3471ada7468..75a972ebf308ffdfd700bc307c48efc75cf07994 100644 --- a/checkpoint-426/optimizer.pt +++ b/checkpoint-426/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b6550e976738f22f913650571284f8eb12dea30c8877134bbe84a1ebd7a73b5e +oid sha256:2bfbb9842afd0e0e7c548c18e2442c293c57767268f19e3c10c10c38c6745a66 size 189552570 diff --git a/checkpoint-426/scheduler.pt b/checkpoint-426/scheduler.pt index 5848057b7abfc3bb9ee3960b26437b6b6f868d53..f519d549a8b824e4f456fcc93b124db502c580e0 100644 --- a/checkpoint-426/scheduler.pt +++ b/checkpoint-426/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1894ac3a34350cc065802675cae4d32438918a7d772edc033390240aaf5d8841 +oid sha256:4d495ecbbc4c3121c1feab1c93a2d37084137f7007f1d94685663700a6993c70 size 1064 diff --git a/checkpoint-426/trainer_state.json b/checkpoint-426/trainer_state.json index c4a5bf78e085f4db18734c660539b2fb3cd0231c..7f3dcc9a3d35c0e738cfef6fb61f1af27db6f86b 100644 --- a/checkpoint-426/trainer_state.json +++ b/checkpoint-426/trainer_state.json @@ -10,69 +10,54 @@ "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, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, "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, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, "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, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, "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, + "eval_accuracy": 0.9897435897435898, + "eval_loss": 0.04228556528687477, + "eval_runtime": 15.2931, + "eval_samples_per_second": 89.256, + "eval_steps_per_second": 11.181, "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, + "eval_loss": 0.044283464550971985, + "eval_runtime": 15.5546, + "eval_samples_per_second": 87.756, + "eval_steps_per_second": 10.994, "step": 426 } ], "logging_steps": 500, - "max_steps": 595, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 7, + "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { diff --git a/checkpoint-426/training_args.bin b/checkpoint-426/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/checkpoint-426/training_args.bin +++ b/checkpoint-426/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240 diff --git a/checkpoint-512/model.safetensors b/checkpoint-512/model.safetensors index c55c352473b4cc1a0d0d29c34da7de2547a9780d..6adabfcc62349df6737e9d27b1eea1db6a28242e 100644 --- a/checkpoint-512/model.safetensors +++ b/checkpoint-512/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6beddb5fef4d20200cfcf763c9460267deb7ad4514b1fa760fa4f89d398502cf +oid sha256:3dc7074a340619ed7e6d2c55347500f679c4e5f5a30b51392562feb760a9f1c7 size 94763496 diff --git a/checkpoint-512/optimizer.pt b/checkpoint-512/optimizer.pt index 1ed40a1fc943647adf5af6cd6c38af7274bb7e5f..e1679b4725d4c67c1b97ee50a0435eac1ef26c67 100644 --- a/checkpoint-512/optimizer.pt +++ b/checkpoint-512/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:2cfb7299eb13d9980cb44dcbfcf90cc3e4a95d2c84fd840b959005ca0b7566d6 +oid sha256:02b1c42643e3e59059b3734bb2a70cedcdb1a3f2b1cfd13f8b4f6444d7bb6e94 size 189552570 diff --git a/checkpoint-512/scheduler.pt b/checkpoint-512/scheduler.pt index 210d846be25edddf75a04893d20df8827db68826..703c196fad30c816082cf754526114a4cef79d68 100644 --- a/checkpoint-512/scheduler.pt +++ b/checkpoint-512/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:7900f26d6583fa63e5c081f39e0457bc3033e9e087a3f2085f881f9fac700fb3 +oid sha256:11fe4216d2ecf2b0e2a833d51c97781e3fccf9e2a48a3b58142f2cf031047fa3 size 1064 diff --git a/checkpoint-512/trainer_state.json b/checkpoint-512/trainer_state.json index e9ac03a7719fa611c37c129c2edb4298bcc35c1b..6facb4fedd01f96108d46988deed6fc7ec95797f 100644 --- a/checkpoint-512/trainer_state.json +++ b/checkpoint-512/trainer_state.json @@ -1,6 +1,6 @@ { - "best_metric": 0.991941391941392, - "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-426", + "best_metric": 0.9926739926739927, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-512", "epoch": 5.997071742313324, "eval_steps": 500, "global_step": 512, @@ -10,88 +10,70 @@ "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, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, "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, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, "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, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, "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, + "eval_accuracy": 0.9897435897435898, + "eval_loss": 0.04228556528687477, + "eval_runtime": 15.2931, + "eval_samples_per_second": 89.256, + "eval_steps_per_second": 11.181, "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, + "eval_loss": 0.044283464550971985, + "eval_runtime": 15.5546, + "eval_samples_per_second": 87.756, + "eval_steps_per_second": 10.994, "step": 426 }, { "epoch": 5.856515373352855, - "grad_norm": 0.01621050015091896, - "learning_rate": 6.1795711069424666e-06, - "loss": 0.053, + "grad_norm": 0.008579956367611885, + "learning_rate": 3.639014166001028e-05, + "loss": 0.055, "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, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04338795691728592, + "eval_runtime": 15.1048, + "eval_samples_per_second": 90.369, + "eval_steps_per_second": 11.321, "step": 512 } ], "logging_steps": 500, - "max_steps": 595, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 7, + "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { diff --git a/checkpoint-512/training_args.bin b/checkpoint-512/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/checkpoint-512/training_args.bin +++ b/checkpoint-512/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240 diff --git a/checkpoint-597/config.json b/checkpoint-597/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-597/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-597/model.safetensors b/checkpoint-597/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..058d139f47aebef271e9f2f7538eaca7a1396a92 --- /dev/null +++ b/checkpoint-597/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15156ce7496feffe08a94b0b76ed04e2398b912b56dfe3d329336cae4f32b8ab +size 94763496 diff --git a/checkpoint-597/optimizer.pt b/checkpoint-597/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..412b6d13c76b0137daf9a72e04b32a00949ba384 --- /dev/null +++ b/checkpoint-597/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14a1d6ff155620819b854ea259f85accf2532818e85d714ea4a614a3b91d42cb +size 189552570 diff --git a/checkpoint-597/rng_state.pth b/checkpoint-597/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..b75b03a09274f8b2c9d5df6328444326ff152635 --- /dev/null +++ b/checkpoint-597/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc9faf0d148cc716f365ebe8d8995bc864d6ff1310fd1b0cf0abdbc5a6b27134 +size 14308 diff --git a/checkpoint-597/scheduler.pt b/checkpoint-597/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..d8edf894f8452030b3eeeb62759bc959cb2bf4bb --- /dev/null +++ b/checkpoint-597/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:08eb49f7720b737aa4a97dd1c15d069bd20c33dc2f3de8105e499906c629b0bb +size 1064 diff --git a/checkpoint-597/trainer_state.json b/checkpoint-597/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..b32c1c325cb4d64af88064e0c91f7d140c35e54d --- /dev/null +++ b/checkpoint-597/trainer_state.json @@ -0,0 +1,112 @@ +{ + "best_metric": 0.9926739926739927, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-512", + "epoch": 6.992679355783309, + "eval_steps": 500, + "global_step": 597, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9897435897435898, + "eval_loss": 0.04228556528687477, + "eval_runtime": 15.2931, + "eval_samples_per_second": 89.256, + "eval_steps_per_second": 11.181, + "step": 341 + }, + { + "epoch": 4.989751098096632, + "eval_accuracy": 0.991941391941392, + "eval_loss": 0.044283464550971985, + "eval_runtime": 15.5546, + "eval_samples_per_second": 87.756, + "eval_steps_per_second": 10.994, + "step": 426 + }, + { + "epoch": 5.856515373352855, + "grad_norm": 0.008579956367611885, + "learning_rate": 3.639014166001028e-05, + "loss": 0.055, + "step": 500 + }, + { + "epoch": 5.997071742313324, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04338795691728592, + "eval_runtime": 15.1048, + "eval_samples_per_second": 90.369, + "eval_steps_per_second": 11.321, + "step": 512 + }, + { + "epoch": 6.992679355783309, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04399614781141281, + "eval_runtime": 15.222, + "eval_samples_per_second": 89.673, + "eval_steps_per_second": 11.234, + "step": 597 + } + ], + "logging_steps": 500, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "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": 8.691905319552e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-597/training_args.bin b/checkpoint-597/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..4594fc4b444b63714f417aa87596ddad831439e5 --- /dev/null +++ b/checkpoint-597/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 +size 5240 diff --git a/checkpoint-683/config.json b/checkpoint-683/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-683/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-683/model.safetensors b/checkpoint-683/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..76ff8f4962f83077487c86d992a40f3b5c77216d --- /dev/null +++ b/checkpoint-683/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52ec3a122dd4e1c3f5b4ae3dc2e2256f512a563d42eb82426ac3f8094e0c4490 +size 94763496 diff --git a/checkpoint-683/optimizer.pt b/checkpoint-683/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..c5e5c8cbd55e730746df23bff2f60d84585bf836 --- /dev/null +++ b/checkpoint-683/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6591188ba0c4025fcbff86219e7c2755b85671c8b4d10202fa03cef2f8831a1 +size 189552570 diff --git a/checkpoint-683/rng_state.pth b/checkpoint-683/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..4e183a55fc3de90d9e729c88460e6879bf9aeb18 --- /dev/null +++ b/checkpoint-683/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43a6b5c1026053dbddf0437264821c51fb9b6be038e7765f49aa97ac2f5521f2 +size 14308 diff --git a/checkpoint-683/scheduler.pt b/checkpoint-683/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4d01af1fdde6c62ebd7697d31558faa49bd9d0bc --- /dev/null +++ b/checkpoint-683/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b712ee88015e952bb23b1f5495f272411d9c19a17334b27df15dc1ebbac98f35 +size 1064 diff --git a/checkpoint-683/trainer_state.json b/checkpoint-683/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..9128eb18e16c02deac3d22cb4285ada55a7bb5c7 --- /dev/null +++ b/checkpoint-683/trainer_state.json @@ -0,0 +1,121 @@ +{ + "best_metric": 0.9926739926739927, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-512", + "epoch": 8.0, + "eval_steps": 500, + "global_step": 683, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9897435897435898, + "eval_loss": 0.04228556528687477, + "eval_runtime": 15.2931, + "eval_samples_per_second": 89.256, + "eval_steps_per_second": 11.181, + "step": 341 + }, + { + "epoch": 4.989751098096632, + "eval_accuracy": 0.991941391941392, + "eval_loss": 0.044283464550971985, + "eval_runtime": 15.5546, + "eval_samples_per_second": 87.756, + "eval_steps_per_second": 10.994, + "step": 426 + }, + { + "epoch": 5.856515373352855, + "grad_norm": 0.008579956367611885, + "learning_rate": 3.639014166001028e-05, + "loss": 0.055, + "step": 500 + }, + { + "epoch": 5.997071742313324, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04338795691728592, + "eval_runtime": 15.1048, + "eval_samples_per_second": 90.369, + "eval_steps_per_second": 11.321, + "step": 512 + }, + { + "epoch": 6.992679355783309, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04399614781141281, + "eval_runtime": 15.222, + "eval_samples_per_second": 89.673, + "eval_steps_per_second": 11.234, + "step": 597 + }, + { + "epoch": 8.0, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.045955102890729904, + "eval_runtime": 15.0695, + "eval_samples_per_second": 90.58, + "eval_steps_per_second": 11.347, + "step": 683 + } + ], + "logging_steps": 500, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "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": 9.933606079488e+16, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-683/training_args.bin b/checkpoint-683/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..4594fc4b444b63714f417aa87596ddad831439e5 --- /dev/null +++ b/checkpoint-683/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 +size 5240 diff --git a/checkpoint-768/config.json b/checkpoint-768/config.json new file mode 100644 index 0000000000000000000000000000000000000000..4550a22abfa803293b5dc6b2633d0201ac2ff44d --- /dev/null +++ b/checkpoint-768/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-768/model.safetensors b/checkpoint-768/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b2519b9c8a86a0f16a98d310dd372f04edd055b7 --- /dev/null +++ b/checkpoint-768/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e5671ffc6960b29ad80f820b4317ec548ff0deefca2464ad359c421c360e64f +size 94763496 diff --git a/checkpoint-768/optimizer.pt b/checkpoint-768/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..b033c12db74cd598ec9bf76fb199a48a614fbda2 --- /dev/null +++ b/checkpoint-768/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7516d91f602bba3efff25344b03b35e85d41b8ce4db2a56619d6dcf6cd2cc2df +size 189552570 diff --git a/checkpoint-768/rng_state.pth b/checkpoint-768/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..6d313342c10e2e661377fbc3297ccd03ea801f26 --- /dev/null +++ b/checkpoint-768/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be73eba9bf4e828d2acb6ab02c40fd90fe2c23584b8cf78d262c1f935416d2e6 +size 14308 diff --git a/checkpoint-768/scheduler.pt b/checkpoint-768/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..9bde14cf908ff3342ecb27c3673a3de161412e6e --- /dev/null +++ b/checkpoint-768/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46a9a21b9ea0c395732aadaa7b885d80eaa4d81c9d9d1eea099d6180e939da87 +size 1064 diff --git a/checkpoint-768/trainer_state.json b/checkpoint-768/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..ecbc324b9a2bf2741b79946abfe8e784ba268475 --- /dev/null +++ b/checkpoint-768/trainer_state.json @@ -0,0 +1,130 @@ +{ + "best_metric": 0.9926739926739927, + "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-512", + "epoch": 8.995607613469986, + "eval_steps": 500, + "global_step": 768, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.9956076134699854, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, + "step": 85 + }, + { + "epoch": 1.9912152269399708, + "eval_accuracy": 0.9860805860805861, + "eval_loss": 0.04661192744970322, + "eval_runtime": 15.611, + "eval_samples_per_second": 87.438, + "eval_steps_per_second": 10.954, + "step": 170 + }, + { + "epoch": 2.998535871156662, + "eval_accuracy": 0.9853479853479854, + "eval_loss": 0.04894961416721344, + "eval_runtime": 15.1886, + "eval_samples_per_second": 89.87, + "eval_steps_per_second": 11.258, + "step": 256 + }, + { + "epoch": 3.994143484626647, + "eval_accuracy": 0.9897435897435898, + "eval_loss": 0.04228556528687477, + "eval_runtime": 15.2931, + "eval_samples_per_second": 89.256, + "eval_steps_per_second": 11.181, + "step": 341 + }, + { + "epoch": 4.989751098096632, + "eval_accuracy": 0.991941391941392, + "eval_loss": 0.044283464550971985, + "eval_runtime": 15.5546, + "eval_samples_per_second": 87.756, + "eval_steps_per_second": 10.994, + "step": 426 + }, + { + "epoch": 5.856515373352855, + "grad_norm": 0.008579956367611885, + "learning_rate": 3.639014166001028e-05, + "loss": 0.055, + "step": 500 + }, + { + "epoch": 5.997071742313324, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04338795691728592, + "eval_runtime": 15.1048, + "eval_samples_per_second": 90.369, + "eval_steps_per_second": 11.321, + "step": 512 + }, + { + "epoch": 6.992679355783309, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04399614781141281, + "eval_runtime": 15.222, + "eval_samples_per_second": 89.673, + "eval_steps_per_second": 11.234, + "step": 597 + }, + { + "epoch": 8.0, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.045955102890729904, + "eval_runtime": 15.0695, + "eval_samples_per_second": 90.58, + "eval_steps_per_second": 11.347, + "step": 683 + }, + { + "epoch": 8.995607613469986, + "eval_accuracy": 0.9926739926739927, + "eval_loss": 0.04587433487176895, + "eval_runtime": 15.048, + "eval_samples_per_second": 90.71, + "eval_steps_per_second": 11.364, + "step": 768 + } + ], + "logging_steps": 500, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "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": 1.1175306839424e+17, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-768/training_args.bin b/checkpoint-768/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..4594fc4b444b63714f417aa87596ddad831439e5 --- /dev/null +++ b/checkpoint-768/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 +size 5240 diff --git a/checkpoint-85/model.safetensors b/checkpoint-85/model.safetensors index a4a245229c8de700b413ffe10c8d03158e50ed5d..5978de9084af52aa127466906b591a88fb3972b7 100644 --- a/checkpoint-85/model.safetensors +++ b/checkpoint-85/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:687e267a3c37d18cce5591d5301b06a4193f540a3ed0c054e9913e5e610aef05 +oid sha256:f9779c7119f4a2210495b9b5237a5030e7c0eddabdde1b1ca2c8239a751959e1 size 94763496 diff --git a/checkpoint-85/optimizer.pt b/checkpoint-85/optimizer.pt index e980b6c8ea2b4bc57a62f661e0cf03de0369051b..00be1771b8f710fd9f603397174f0c1d9b76e896 100644 --- a/checkpoint-85/optimizer.pt +++ b/checkpoint-85/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:8ff974afccdd073da4ec78806e9470c99c469ae8a871771439016949486039a9 +oid sha256:4a6690ddab61f45db23b28120bb5c39be4f9a152f1c88b3ae432902f89efc452 size 189552570 diff --git a/checkpoint-85/scheduler.pt b/checkpoint-85/scheduler.pt index fbad7a97a88117443a8a9d272b29a6c604cb6f57..34476a2d3ba89f1cfa273fb0e782857bad8e531f 100644 --- a/checkpoint-85/scheduler.pt +++ b/checkpoint-85/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:daf90af7d8b581c8b4882d01681a4f4fb886f79b4c5624a5d256aba635f56702 +oid sha256:324ecccddc3afacd1a2ab429e35f5e62f986b75e609998929bd8ab5fcbb2e6a7 size 1064 diff --git a/checkpoint-85/trainer_state.json b/checkpoint-85/trainer_state.json index c48612280f108c8e33785720e1aab136b737ec2d..0b04ec572211545e37724e82f5b03d5edc26df44 100644 --- a/checkpoint-85/trainer_state.json +++ b/checkpoint-85/trainer_state.json @@ -1,5 +1,5 @@ { - "best_metric": 0.9787545787545787, + "best_metric": 0.9772893772893773, "best_model_checkpoint": "distilhubert-finetuned-cry-detector/checkpoint-85", "epoch": 0.9956076134699854, "eval_steps": 500, @@ -10,21 +10,18 @@ "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, + "eval_accuracy": 0.9772893772893773, + "eval_loss": 0.06922276318073273, + "eval_runtime": 15.3212, + "eval_samples_per_second": 89.092, + "eval_steps_per_second": 11.161, "step": 85 } ], "logging_steps": 500, - "max_steps": 595, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 7, + "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { diff --git a/checkpoint-85/training_args.bin b/checkpoint-85/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/checkpoint-85/training_args.bin +++ b/checkpoint-85/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240 diff --git a/model.safetensors b/model.safetensors index b47cf29d4974efcc2d310aa700ba4749b22632c6..6adabfcc62349df6737e9d27b1eea1db6a28242e 100644 --- a/model.safetensors +++ b/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:94760f7df9dc2c6bad702593ec0871720fa6ae352e88f79c71bcfccc8fab1a09 +oid sha256:3dc7074a340619ed7e6d2c55347500f679c4e5f5a30b51392562feb760a9f1c7 size 94763496 diff --git a/runs/Sep02_11-59-10_ubumarcos/events.out.tfevents.1725271162.ubumarcos b/runs/Sep02_11-59-10_ubumarcos/events.out.tfevents.1725271162.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..9696cf6e69ef726c0280ed961143d68714f7d655 --- /dev/null +++ b/runs/Sep02_11-59-10_ubumarcos/events.out.tfevents.1725271162.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:daedd80eba2df9bf6e585e75b20fd56f783d063e9e0176793a1ccb5efbad4648 +size 4181 diff --git a/runs/Sep02_12-01-00_ubumarcos/events.out.tfevents.1725271267.ubumarcos b/runs/Sep02_12-01-00_ubumarcos/events.out.tfevents.1725271267.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..aee11a10cebb30cfd55743232b967a5c6608407b --- /dev/null +++ b/runs/Sep02_12-01-00_ubumarcos/events.out.tfevents.1725271267.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3791f55fc98e6686de02502757f77c72ad5103e107fe24f6a814ff94a9ae8929 +size 4181 diff --git a/runs/Sep02_13-43-36_ubumarcos/events.out.tfevents.1725277433.ubumarcos b/runs/Sep02_13-43-36_ubumarcos/events.out.tfevents.1725277433.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..c4e34b5bde5e12a3e005db538faae5fd6950ae25 --- /dev/null +++ b/runs/Sep02_13-43-36_ubumarcos/events.out.tfevents.1725277433.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a3e4c6e9353c0948cbc18284a8292cef0871de8232c7156ff4357500e6a0687 +size 4181 diff --git a/runs/Sep02_13-48-37_ubumarcos/events.out.tfevents.1725277727.ubumarcos b/runs/Sep02_13-48-37_ubumarcos/events.out.tfevents.1725277727.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..5b6f775b9345d2f25d9ef03090f142ad47d082d6 --- /dev/null +++ b/runs/Sep02_13-48-37_ubumarcos/events.out.tfevents.1725277727.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59f73d8d9e35663fcd1c9760a8c3ae31879ee7a6adacee82f10c083c226d4b64 +size 6620 diff --git a/runs/Sep02_13-56-21_ubumarcos/events.out.tfevents.1725278196.ubumarcos b/runs/Sep02_13-56-21_ubumarcos/events.out.tfevents.1725278196.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..b118f8059183cfc98e687e870e8719fd737d8516 --- /dev/null +++ b/runs/Sep02_13-56-21_ubumarcos/events.out.tfevents.1725278196.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7426d798e5d4bc67124cf4249b1e0f71dfa5baa3049b5b41082e8855c19460d5 +size 4181 diff --git a/runs/Sep02_13-56-50_ubumarcos/events.out.tfevents.1725278220.ubumarcos b/runs/Sep02_13-56-50_ubumarcos/events.out.tfevents.1725278220.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..e7a95cbce0c1f39bf18f63bf6577eaac37c97f89 --- /dev/null +++ b/runs/Sep02_13-56-50_ubumarcos/events.out.tfevents.1725278220.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d30c72fff704fb32f50bd175886ac62d1cccde2e596cc296dce458a598e0fb46 +size 4181 diff --git a/runs/Sep02_13-57-15_ubumarcos/events.out.tfevents.1725278242.ubumarcos b/runs/Sep02_13-57-15_ubumarcos/events.out.tfevents.1725278242.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..d853a7a6ed962e34f073daae537f8e2553f72b80 --- /dev/null +++ b/runs/Sep02_13-57-15_ubumarcos/events.out.tfevents.1725278242.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9bba35670df0bc288f886dedea274c70770d3919b50788c7769399bd79a3cf07 +size 4181 diff --git a/runs/Sep03_13-17-36_ubumarcos/events.out.tfevents.1725362270.ubumarcos b/runs/Sep03_13-17-36_ubumarcos/events.out.tfevents.1725362270.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..927e884810e50a9f02d23ec41bd7c92fbb376368 --- /dev/null +++ b/runs/Sep03_13-17-36_ubumarcos/events.out.tfevents.1725362270.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05b3078a3240b22c50af6c685d84d3be713e1d822e0a841cdcc84813fec82411 +size 4181 diff --git a/runs/Sep03_16-02-29_ubumarcos/events.out.tfevents.1725372161.ubumarcos b/runs/Sep03_16-02-29_ubumarcos/events.out.tfevents.1725372161.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..5f8042de938c88be11af6c4087f8ee5872baddd0 --- /dev/null +++ b/runs/Sep03_16-02-29_ubumarcos/events.out.tfevents.1725372161.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92a509bedb440391d219bbf0822727ea85a924404969fd1848f508b8aa9e5f3a +size 6474 diff --git a/runs/Sep03_16-11-40_ubumarcos/events.out.tfevents.1725372708.ubumarcos b/runs/Sep03_16-11-40_ubumarcos/events.out.tfevents.1725372708.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..c77c29a5d7bf299d4102d9823be59ca272b6d027 --- /dev/null +++ b/runs/Sep03_16-11-40_ubumarcos/events.out.tfevents.1725372708.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5264720d0752d2305d43355e9021a82c4ffbf1ba043ecda323a6c23dda01549f +size 5810 diff --git a/runs/Sep03_16-12-14_ubumarcos/events.out.tfevents.1725372743.ubumarcos b/runs/Sep03_16-12-14_ubumarcos/events.out.tfevents.1725372743.ubumarcos new file mode 100644 index 0000000000000000000000000000000000000000..4617ba6f79d7387cfc61d93ec3a9df82de214a87 --- /dev/null +++ b/runs/Sep03_16-12-14_ubumarcos/events.out.tfevents.1725372743.ubumarcos @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30b82d767d9b517c51ab37d940293e2436994c8b800fc426d8232d6db2e34c8c +size 9276 diff --git a/training_args.bin b/training_args.bin index 8862d582702297ee422932b621da79a72d7db40c..4594fc4b444b63714f417aa87596ddad831439e5 100644 --- a/training_args.bin +++ b/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:caa3df7f2810faa9c35a46d178b12337edb0b134aa27dc1689b302d9bcc9d3d0 +oid sha256:abb08b4365c154a27d61d8cdfad62c4717bf003430323e6c3916f01c3121d721 size 5240