diff --git a/checkpoint-100/config.json b/checkpoint-100/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-100/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-100/generation_config.json b/checkpoint-100/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-100/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-100/model.safetensors b/checkpoint-100/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6f16463b0e9111e0553cd672bfcf0aad71d896e9 --- /dev/null +++ b/checkpoint-100/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:657c075db432a88a3f9904c83183b312d2d62c7593c49e6ff59061f775381796 +size 966995080 diff --git a/checkpoint-100/optimizer.pt b/checkpoint-100/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..d762f12a11b79eb1c19d2ab0916ddd5ef2f4c388 --- /dev/null +++ b/checkpoint-100/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb528f0a4edb75abac73e8218838a7cbc024cadebb3b4d6f30bcce2305919f3d +size 1925064044 diff --git a/checkpoint-100/preprocessor_config.json b/checkpoint-100/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-100/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-100/rng_state.pth b/checkpoint-100/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..5496b7ddd9dc2c64296d3f176b8a29a80396e434 --- /dev/null +++ b/checkpoint-100/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:56891aaf8ac3e28df502c97b7baca2742be4c09167d04576400c329248b881d2 +size 14244 diff --git a/checkpoint-100/scaler.pt b/checkpoint-100/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f6b3b7531b67d6a3286c705af0b9b33aed1a3dd8 --- /dev/null +++ b/checkpoint-100/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac4a558c5b93581a5c41e2922404490319bcd15a10296c927a72a41012ff7f27 +size 988 diff --git a/checkpoint-100/scheduler.pt b/checkpoint-100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..552edc35efaa1f09a3bebbc4f69162427f81c15f --- /dev/null +++ b/checkpoint-100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e3f011953770dc18a51194e919361544c1e41f047fac26af2e17015642e43c7d +size 1064 diff --git a/checkpoint-100/trainer_state.json b/checkpoint-100/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..f38fda4c21d01c69389677c4cb787ef4d00ccb89 --- /dev/null +++ b/checkpoint-100/trainer_state.json @@ -0,0 +1,78 @@ +{ + "best_global_step": 100, + "best_metric": 36.6100851831126, + "best_model_checkpoint": "./JUDIC/checkpoint-100", + "epoch": 1.1764705882352942, + "eval_steps": 100, + "global_step": 100, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 9.23473281024e+17, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-100/training_args.bin b/checkpoint-100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432 diff --git a/checkpoint-200/model.safetensors b/checkpoint-200/model.safetensors index 343b0a8469640b553a0d263210d6e14edbcb29a0..3bd68a35a39d09a430b54fed584c2195eeb72b45 100644 --- a/checkpoint-200/model.safetensors +++ b/checkpoint-200/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6b2892e2e8a768289512d80a4527ba6a28e97b8786cb115dc57ba64fb5dd1b35 +oid sha256:6f81c82d6052baacb7e66c24e8c708a630cd8cedf72c4ff7a9ec6f231ecd905d size 966995080 diff --git a/checkpoint-200/optimizer.pt b/checkpoint-200/optimizer.pt index 1162650b01b4ca2b3107cb07cce276897d97cb1f..d20be6c1b966b58f32fd94de2b811baa760d6cce 100644 --- a/checkpoint-200/optimizer.pt +++ b/checkpoint-200/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1fee27de3f002951623132d14fb9111b69495b5382d4bd6cfe5b785e99689855 +oid sha256:38b02504b8786f500125ee911e1e4fe3881605e2517b98180d456037a54f5715 size 1925064044 diff --git a/checkpoint-200/rng_state.pth b/checkpoint-200/rng_state.pth index fe4f933b2658d123b9b67d8892afb838a7d2298c..43ea5e43091815821b43f5bdf6190db69a4e1d01 100644 --- a/checkpoint-200/rng_state.pth +++ b/checkpoint-200/rng_state.pth @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:92ade350e84ed8edc73f8df3ecb0b81d4efe23f223f59b19ab9f901c5bd3f39f +oid sha256:3b7e1937d511f9c3420c5c21f44a372adbc882c5626121f8a8c3d5d9e8ade15b size 14244 diff --git a/checkpoint-200/scaler.pt b/checkpoint-200/scaler.pt index be7ee849707316eadd55a5ea77263caf4154cd7a..5e6e0ae55d862fb176c30540286ca98e17281de3 100644 --- a/checkpoint-200/scaler.pt +++ b/checkpoint-200/scaler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:210384cc8aac9b5f42f3d07f3f98c31fe1102ba37532cf4f431d4379f4392fbb +oid sha256:aa75768e65458cb8b0fa3d934398492a1aa75cd3ae318e75b37395db60dc10a2 size 988 diff --git a/checkpoint-200/scheduler.pt b/checkpoint-200/scheduler.pt index 635dafb37d95cb360b0283b33e79a1ee49a29ff7..3052de27245494d518841af4073d5a6ddeb10b0b 100644 --- a/checkpoint-200/scheduler.pt +++ b/checkpoint-200/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:8b551e44d36a93f06118444de4d46a48d68025b941ad3f20a21f7cd1e45875cd +oid sha256:23427674df6bb33784b590ae4aa53899078e927f5e43e1c513019f2bf257b7de size 1064 diff --git a/checkpoint-200/trainer_state.json b/checkpoint-200/trainer_state.json index 1db98243bac1b0f50dc358f3727c1c93ccbccebf..4eb15e70809f25c802ac4121610dc71d70a90931 100644 --- a/checkpoint-200/trainer_state.json +++ b/checkpoint-200/trainer_state.json @@ -1,9 +1,9 @@ { "best_global_step": 200, - "best_metric": 41.783249393769815, + "best_metric": 29.938444320089534, "best_model_checkpoint": "./JUDIC/checkpoint-200", "epoch": 2.3529411764705883, - "eval_steps": 200, + "eval_steps": 100, "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, @@ -11,89 +11,98 @@ "log_history": [ { "epoch": 0.23529411764705882, - "grad_norm": 39.55274200439453, - "learning_rate": 8.000000000000001e-07, - "loss": 2.8704, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, "step": 20 }, { "epoch": 0.47058823529411764, - "grad_norm": 16.861303329467773, - "learning_rate": 1.8000000000000001e-06, - "loss": 2.4422, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, "step": 40 }, { "epoch": 0.7058823529411765, - "grad_norm": 10.871281623840332, - "learning_rate": 2.8000000000000003e-06, - "loss": 1.9044, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, "step": 60 }, { "epoch": 0.9411764705882353, - "grad_norm": 7.858098030090332, - "learning_rate": 3.8000000000000005e-06, - "loss": 1.4465, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, "step": 80 }, { "epoch": 1.1764705882352942, - "grad_norm": 7.4879231452941895, - "learning_rate": 4.800000000000001e-06, - "loss": 1.3179, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, "step": 100 }, { "epoch": 1.4117647058823528, - "grad_norm": 7.5675835609436035, - "learning_rate": 5.8e-06, - "loss": 1.0743, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, "step": 120 }, { "epoch": 1.6470588235294117, - "grad_norm": 6.898285865783691, - "learning_rate": 6.800000000000001e-06, - "loss": 1.0679, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, "step": 140 }, { "epoch": 1.8823529411764706, - "grad_norm": 7.592809677124023, - "learning_rate": 7.800000000000002e-06, - "loss": 1.0381, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, "step": 160 }, { "epoch": 2.1176470588235294, - "grad_norm": 8.085546493530273, - "learning_rate": 8.8e-06, - "loss": 0.8211, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, "step": 180 }, { "epoch": 2.3529411764705883, - "grad_norm": 5.359922885894775, - "learning_rate": 9.800000000000001e-06, - "loss": 0.7182, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, "step": 200 }, { "epoch": 2.3529411764705883, - "eval_loss": 0.6759204864501953, - "eval_runtime": 326.8541, - "eval_samples_per_second": 2.083, - "eval_steps_per_second": 0.263, - "eval_wer": 41.783249393769815, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, "step": 200 } ], "logging_steps": 20, - "max_steps": 425, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 5, - "save_steps": 200, + "num_train_epochs": 10, + "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { diff --git a/checkpoint-200/training_args.bin b/checkpoint-200/training_args.bin index 2fb50261b2b138eb3f5bc201f0fbd9ad4e76cedc..dda69251eaface4172ea0a644fb661871a9152e7 100644 --- a/checkpoint-200/training_args.bin +++ b/checkpoint-200/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a8560064d3c95b83f5ff13cbbf3f02ecfc5b06383d6bc753a071d8b592db8d83 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d size 5432 diff --git a/checkpoint-300/config.json b/checkpoint-300/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-300/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-300/generation_config.json b/checkpoint-300/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-300/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-300/model.safetensors b/checkpoint-300/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a7452119b870d552548ec14f838796b0214a5eff --- /dev/null +++ b/checkpoint-300/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6fea0b87575944e3fad7d172cfb03dc2c5fbd22d6556d834ff0a0b0568706ad +size 966995080 diff --git a/checkpoint-300/optimizer.pt b/checkpoint-300/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..abab6f49775e2b7a273c574bb20ebbc73db0ff49 --- /dev/null +++ b/checkpoint-300/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd683dced7bbbadf49aeaca0d526ec98fa5ac9eb42ff2852ce2c726b56120616 +size 1925064044 diff --git a/checkpoint-300/preprocessor_config.json b/checkpoint-300/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-300/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-300/rng_state.pth b/checkpoint-300/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..f900ff810e48bc68c7112cd16e72b5fdab13d96e --- /dev/null +++ b/checkpoint-300/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c75de8aef036ad6437dee31b4cc33fdaf4d9234973688095b81d98166c4ca11 +size 14244 diff --git a/checkpoint-300/scaler.pt b/checkpoint-300/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..3411b3be3ab301be42b18f92d1a92b2a7546da17 --- /dev/null +++ b/checkpoint-300/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1f55124c020962c0cda07fe10a52f1e374d6480f4931ceea4539f39f03b1ff2e +size 988 diff --git a/checkpoint-300/scheduler.pt b/checkpoint-300/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..b4271c1ae2d5236f56366876f5c950fd6e43b2c4 --- /dev/null +++ b/checkpoint-300/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9122dc6585a724b33eebfaf95afd97b180142b91a30efe1fe1ddc84cb2148c14 +size 1064 diff --git a/checkpoint-300/trainer_state.json b/checkpoint-300/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..c0874d362fe400a9c6a10bc873a816dbf4fe28ca --- /dev/null +++ b/checkpoint-300/trainer_state.json @@ -0,0 +1,166 @@ +{ + "best_global_step": 200, + "best_metric": 29.938444320089534, + "best_model_checkpoint": "./JUDIC/checkpoint-200", + "epoch": 3.5294117647058822, + "eval_steps": 100, + "global_step": 300, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + }, + { + "epoch": 1.4117647058823528, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, + "step": 120 + }, + { + "epoch": 1.6470588235294117, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, + "step": 140 + }, + { + "epoch": 1.8823529411764706, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, + "step": 160 + }, + { + "epoch": 2.1176470588235294, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, + "step": 180 + }, + { + "epoch": 2.3529411764705883, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, + "step": 200 + }, + { + "epoch": 2.3529411764705883, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, + "step": 200 + }, + { + "epoch": 2.588235294117647, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, + "step": 220 + }, + { + "epoch": 2.8235294117647056, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, + "step": 240 + }, + { + "epoch": 3.0588235294117645, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, + "step": 260 + }, + { + "epoch": 3.2941176470588234, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, + "step": 280 + }, + { + "epoch": 3.5294117647058822, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, + "step": 300 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2.770419843072e+18, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-300/training_args.bin b/checkpoint-300/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-300/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432 diff --git a/checkpoint-400/model.safetensors b/checkpoint-400/model.safetensors index 94a5fe0e54a144b57535af593073cbdee6ce9412..e4276373a161734bd2d5ea87e7745ef0989c5ca2 100644 --- a/checkpoint-400/model.safetensors +++ b/checkpoint-400/model.safetensors @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:2ccf6e0b18e00966cc3aa9593f5bc67a572fe68bf2f1c193e3a9f23842ef250c +oid sha256:826cae9d99f0d4d549aa98fab1b85fc27755f00ce7b56ac7a10cf63a2a6fb584 size 966995080 diff --git a/checkpoint-400/optimizer.pt b/checkpoint-400/optimizer.pt index 3b261534f81f60892e8bc95099246fa1680dc56b..ab2361ea80d504b277c7ae7365a146c11a2ef10f 100644 --- a/checkpoint-400/optimizer.pt +++ b/checkpoint-400/optimizer.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:31770e956f5864ecbeb44af1a8424c7ec91ab93053c2e7fc7840ec0c5aa72442 +oid sha256:f32422f07da705bcb7284e1286b3f99b2f566ecd46a211462f9b454aff8d60a5 size 1925064044 diff --git a/checkpoint-400/rng_state.pth b/checkpoint-400/rng_state.pth index 08eac3c30236aef9604b2841cf09b4c3bbd06ad4..6aa1c5b7cadc31616052952ca6bfba1bccb49c7c 100644 --- a/checkpoint-400/rng_state.pth +++ b/checkpoint-400/rng_state.pth @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:3327069f4314534e1155a5e53b9e4445074a676d823a310baf6fdcb1c7c4b00a +oid sha256:30c47e162e68dbe940a216a2770ed84b7751c4b1e55b4de071db84c5e99904ed size 14244 diff --git a/checkpoint-400/scaler.pt b/checkpoint-400/scaler.pt index b445f888c7aaf82cb99165d7367ba19a005fb68f..491f09774e1027de5ee4ac7d6dfd17df4da7717f 100644 --- a/checkpoint-400/scaler.pt +++ b/checkpoint-400/scaler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:643802dd3517303465f449d0a6e794d1e1488b20ad8fa83264855b4ced649eac +oid sha256:9ba4abf7693b716a8ac4dd2dada1c394abd2cf2ea9e18f42a3a5733573c24b77 size 988 diff --git a/checkpoint-400/scheduler.pt b/checkpoint-400/scheduler.pt index 692f899b2b66576490b7ecf53707f98daba93a77..a054329091489253bc3e2398ff9260c7da2103ec 100644 --- a/checkpoint-400/scheduler.pt +++ b/checkpoint-400/scheduler.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:98c75408be126709cd3a95cfde165c89fc584aa0301e5e1c94cc95f4373b1d92 +oid sha256:79d838619a5e73d7fb3ccfd45f48dd11a5b5251829dba401a7a2d6341f3b7f2d size 1064 diff --git a/checkpoint-400/trainer_state.json b/checkpoint-400/trainer_state.json index 2acf3f2109844a6d275d89ee3cfa896d0874fb32..1d20bf2d07c919d52321ae36bf3aa3c35137d668 100644 --- a/checkpoint-400/trainer_state.json +++ b/checkpoint-400/trainer_state.json @@ -1,9 +1,9 @@ { "best_global_step": 400, - "best_metric": 37.20698874588074, + "best_metric": 25.511409562892496, "best_model_checkpoint": "./JUDIC/checkpoint-400", "epoch": 4.705882352941177, - "eval_steps": 200, + "eval_steps": 100, "global_step": 400, "is_hyper_param_search": false, "is_local_process_zero": true, @@ -11,168 +11,186 @@ "log_history": [ { "epoch": 0.23529411764705882, - "grad_norm": 39.55274200439453, - "learning_rate": 8.000000000000001e-07, - "loss": 2.8704, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, "step": 20 }, { "epoch": 0.47058823529411764, - "grad_norm": 16.861303329467773, - "learning_rate": 1.8000000000000001e-06, - "loss": 2.4422, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, "step": 40 }, { "epoch": 0.7058823529411765, - "grad_norm": 10.871281623840332, - "learning_rate": 2.8000000000000003e-06, - "loss": 1.9044, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, "step": 60 }, { "epoch": 0.9411764705882353, - "grad_norm": 7.858098030090332, - "learning_rate": 3.8000000000000005e-06, - "loss": 1.4465, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, "step": 80 }, { "epoch": 1.1764705882352942, - "grad_norm": 7.4879231452941895, - "learning_rate": 4.800000000000001e-06, - "loss": 1.3179, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, "step": 100 }, { "epoch": 1.4117647058823528, - "grad_norm": 7.5675835609436035, - "learning_rate": 5.8e-06, - "loss": 1.0743, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, "step": 120 }, { "epoch": 1.6470588235294117, - "grad_norm": 6.898285865783691, - "learning_rate": 6.800000000000001e-06, - "loss": 1.0679, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, "step": 140 }, { "epoch": 1.8823529411764706, - "grad_norm": 7.592809677124023, - "learning_rate": 7.800000000000002e-06, - "loss": 1.0381, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, "step": 160 }, { "epoch": 2.1176470588235294, - "grad_norm": 8.085546493530273, - "learning_rate": 8.8e-06, - "loss": 0.8211, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, "step": 180 }, { "epoch": 2.3529411764705883, - "grad_norm": 5.359922885894775, - "learning_rate": 9.800000000000001e-06, - "loss": 0.7182, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, "step": 200 }, { "epoch": 2.3529411764705883, - "eval_loss": 0.6759204864501953, - "eval_runtime": 326.8541, - "eval_samples_per_second": 2.083, - "eval_steps_per_second": 0.263, - "eval_wer": 41.783249393769815, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, "step": 200 }, { "epoch": 2.588235294117647, - "grad_norm": 3.346689462661743, - "learning_rate": 9.28888888888889e-06, - "loss": 0.7775, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, "step": 220 }, { "epoch": 2.8235294117647056, - "grad_norm": 4.038958549499512, - "learning_rate": 8.400000000000001e-06, - "loss": 0.6748, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, "step": 240 }, { "epoch": 3.0588235294117645, - "grad_norm": 3.325453758239746, - "learning_rate": 7.511111111111111e-06, - "loss": 0.66, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, "step": 260 }, { "epoch": 3.2941176470588234, - "grad_norm": 3.325045108795166, - "learning_rate": 6.6222222222222236e-06, - "loss": 0.5344, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, "step": 280 }, { "epoch": 3.5294117647058822, - "grad_norm": 3.9428045749664307, - "learning_rate": 5.733333333333334e-06, - "loss": 0.5228, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, "step": 300 }, { "epoch": 3.764705882352941, - "grad_norm": 3.552807092666626, - "learning_rate": 4.8444444444444446e-06, - "loss": 0.5771, + "grad_norm": 2.77553391456604, + "learning_rate": 7.1066666666666675e-06, + "loss": 0.2342, "step": 320 }, { "epoch": 4.0, - "grad_norm": 4.557582378387451, - "learning_rate": 3.955555555555556e-06, - "loss": 0.4784, + "grad_norm": 3.944244861602783, + "learning_rate": 6.8400000000000014e-06, + "loss": 0.1751, "step": 340 }, { "epoch": 4.235294117647059, - "grad_norm": 4.543866157531738, - "learning_rate": 3.066666666666667e-06, - "loss": 0.4485, + "grad_norm": 6.829845905303955, + "learning_rate": 6.573333333333334e-06, + "loss": 0.1431, "step": 360 }, { "epoch": 4.470588235294118, - "grad_norm": 3.4673140048980713, - "learning_rate": 2.1777777777777777e-06, - "loss": 0.4194, + "grad_norm": 3.4126970767974854, + "learning_rate": 6.3066666666666676e-06, + "loss": 0.1468, "step": 380 }, { "epoch": 4.705882352941177, - "grad_norm": 3.8203773498535156, - "learning_rate": 1.288888888888889e-06, - "loss": 0.3985, + "grad_norm": 3.5233240127563477, + "learning_rate": 6.040000000000001e-06, + "loss": 0.1376, "step": 400 }, { "epoch": 4.705882352941177, - "eval_loss": 0.44990459084510803, - "eval_runtime": 307.4758, + "eval_loss": 0.27411767840385437, + "eval_runtime": 307.4101, "eval_samples_per_second": 2.215, "eval_steps_per_second": 0.28, - "eval_wer": 37.20698874588074, + "eval_wer": 25.511409562892496, "step": 400 } ], "logging_steps": 20, - "max_steps": 425, + "max_steps": 850, "num_input_tokens_seen": 0, - "num_train_epochs": 5, - "save_steps": 200, + "num_train_epochs": 10, + "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { diff --git a/checkpoint-400/training_args.bin b/checkpoint-400/training_args.bin index 2fb50261b2b138eb3f5bc201f0fbd9ad4e76cedc..dda69251eaface4172ea0a644fb661871a9152e7 100644 --- a/checkpoint-400/training_args.bin +++ b/checkpoint-400/training_args.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a8560064d3c95b83f5ff13cbbf3f02ecfc5b06383d6bc753a071d8b592db8d83 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d size 5432 diff --git a/checkpoint-500/config.json b/checkpoint-500/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-500/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-500/generation_config.json b/checkpoint-500/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-500/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-500/model.safetensors b/checkpoint-500/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3baea2dd0a0f9bcbc13fa1b9bf7e311781ef9dee --- /dev/null +++ b/checkpoint-500/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d528589c3533907ce7d62b0f82eb8e4b39324a2ede9668d4637805894ab9c265 +size 966995080 diff --git a/checkpoint-500/optimizer.pt b/checkpoint-500/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..d6b2e6d7aee21e3f3ad124473141da96497b1c7d --- /dev/null +++ b/checkpoint-500/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2312d2bf46c62764990aa1f900a0cc3f4aa4aceee8f2f20998ac84b07c5d89a6 +size 1925064044 diff --git a/checkpoint-500/preprocessor_config.json b/checkpoint-500/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-500/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-500/rng_state.pth b/checkpoint-500/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..12038a0ae33e8b3243ff34d3b50437b2cfc1a471 --- /dev/null +++ b/checkpoint-500/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:825599fa0730d95fb5920cab5c6707dfe544fd8910dfd0d5035f628a04b4f17f +size 14244 diff --git a/checkpoint-500/scaler.pt b/checkpoint-500/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..6962155f1fb80820f03fc0d6bef54cf04f9f8c12 --- /dev/null +++ b/checkpoint-500/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc4ce6f6531c680ba0e80ac9ab7f956e79be869228ce1b03c87da5b41fcf9d77 +size 988 diff --git a/checkpoint-500/scheduler.pt b/checkpoint-500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..d132e36966d34b9978922358defe2e7e71af4665 --- /dev/null +++ b/checkpoint-500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e37d0e3abe4f26a67104e2dc03695cbf6a36428ae4534496e8677a64b279559 +size 1064 diff --git a/checkpoint-500/trainer_state.json b/checkpoint-500/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..11d322e29c9644ba39ee30cbe5d5605bf43426d7 --- /dev/null +++ b/checkpoint-500/trainer_state.json @@ -0,0 +1,254 @@ +{ + "best_global_step": 400, + "best_metric": 25.511409562892496, + "best_model_checkpoint": "./JUDIC/checkpoint-400", + "epoch": 5.882352941176471, + "eval_steps": 100, + "global_step": 500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + }, + { + "epoch": 1.4117647058823528, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, + "step": 120 + }, + { + "epoch": 1.6470588235294117, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, + "step": 140 + }, + { + "epoch": 1.8823529411764706, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, + "step": 160 + }, + { + "epoch": 2.1176470588235294, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, + "step": 180 + }, + { + "epoch": 2.3529411764705883, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, + "step": 200 + }, + { + "epoch": 2.3529411764705883, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, + "step": 200 + }, + { + "epoch": 2.588235294117647, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, + "step": 220 + }, + { + "epoch": 2.8235294117647056, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, + "step": 240 + }, + { + "epoch": 3.0588235294117645, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, + "step": 260 + }, + { + "epoch": 3.2941176470588234, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, + "step": 280 + }, + { + "epoch": 3.5294117647058822, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, + "step": 300 + }, + { + "epoch": 3.764705882352941, + "grad_norm": 2.77553391456604, + "learning_rate": 7.1066666666666675e-06, + "loss": 0.2342, + "step": 320 + }, + { + "epoch": 4.0, + "grad_norm": 3.944244861602783, + "learning_rate": 6.8400000000000014e-06, + "loss": 0.1751, + "step": 340 + }, + { + "epoch": 4.235294117647059, + "grad_norm": 6.829845905303955, + "learning_rate": 6.573333333333334e-06, + "loss": 0.1431, + "step": 360 + }, + { + "epoch": 4.470588235294118, + "grad_norm": 3.4126970767974854, + "learning_rate": 6.3066666666666676e-06, + "loss": 0.1468, + "step": 380 + }, + { + "epoch": 4.705882352941177, + "grad_norm": 3.5233240127563477, + "learning_rate": 6.040000000000001e-06, + "loss": 0.1376, + "step": 400 + }, + { + "epoch": 4.705882352941177, + "eval_loss": 0.27411767840385437, + "eval_runtime": 307.4101, + "eval_samples_per_second": 2.215, + "eval_steps_per_second": 0.28, + "eval_wer": 25.511409562892496, + "step": 400 + }, + { + "epoch": 4.9411764705882355, + "grad_norm": 2.9697437286376953, + "learning_rate": 5.7733333333333345e-06, + "loss": 0.1395, + "step": 420 + }, + { + "epoch": 5.176470588235294, + "grad_norm": 3.495859384536743, + "learning_rate": 5.506666666666667e-06, + "loss": 0.1046, + "step": 440 + }, + { + "epoch": 5.411764705882353, + "grad_norm": 4.7592902183532715, + "learning_rate": 5.240000000000001e-06, + "loss": 0.1179, + "step": 460 + }, + { + "epoch": 5.647058823529412, + "grad_norm": 7.579683303833008, + "learning_rate": 4.973333333333334e-06, + "loss": 0.1361, + "step": 480 + }, + { + "epoch": 5.882352941176471, + "grad_norm": 4.949639797210693, + "learning_rate": 4.706666666666667e-06, + "loss": 0.0935, + "step": 500 + }, + { + "epoch": 5.882352941176471, + "eval_loss": 0.2585192918777466, + "eval_runtime": 321.1171, + "eval_samples_per_second": 2.121, + "eval_steps_per_second": 0.268, + "eval_wer": 26.26997450724367, + "step": 500 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.61736640512e+18, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-500/training_args.bin b/checkpoint-500/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432 diff --git a/checkpoint-600/config.json b/checkpoint-600/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-600/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-600/generation_config.json b/checkpoint-600/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-600/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-600/model.safetensors b/checkpoint-600/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7831da9ab1aaf92bc20217f794e8e07a3b04837b --- /dev/null +++ b/checkpoint-600/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3799349bbd9f1b5bc5b63fe1366f4ff31d0d06cabecdbfa7aef16ea012bf4d1d +size 966995080 diff --git a/checkpoint-600/optimizer.pt b/checkpoint-600/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..e8f44b73b5a7fb941b738dea80993c0c027fcc04 --- /dev/null +++ b/checkpoint-600/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c86ac17197344a70407c4bcbd330f7eb44fbf135c27a3a8744bc30e21c62cf5 +size 1925064044 diff --git a/checkpoint-600/preprocessor_config.json b/checkpoint-600/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-600/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-600/rng_state.pth b/checkpoint-600/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..989f32d0eb3144e63c05a2e31cce91ca0d4f043a --- /dev/null +++ b/checkpoint-600/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:480661d528573adc366803cd6192719058233ab699007a6e9d72a2a4445eab53 +size 14244 diff --git a/checkpoint-600/scaler.pt b/checkpoint-600/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f63028624d944e9a11edcf6f6e36becb9609b044 --- /dev/null +++ b/checkpoint-600/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66ba7d647488928a864725af1fbe0406afc56021a9dd59ba237971070b725cb0 +size 988 diff --git a/checkpoint-600/scheduler.pt b/checkpoint-600/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..395c6d70393eb7a102b378eb393edfd60a464fbf --- /dev/null +++ b/checkpoint-600/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d46dd22904aaf85d0644f49c10885b626ac117fc0ebc918d1c08bcd9bcaed9a +size 1064 diff --git a/checkpoint-600/trainer_state.json b/checkpoint-600/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..91e6a3f315dc50875953239b595a519db2378585 --- /dev/null +++ b/checkpoint-600/trainer_state.json @@ -0,0 +1,298 @@ +{ + "best_global_step": 600, + "best_metric": 24.187029783000686, + "best_model_checkpoint": "./JUDIC/checkpoint-600", + "epoch": 7.0588235294117645, + "eval_steps": 100, + "global_step": 600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + }, + { + "epoch": 1.4117647058823528, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, + "step": 120 + }, + { + "epoch": 1.6470588235294117, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, + "step": 140 + }, + { + "epoch": 1.8823529411764706, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, + "step": 160 + }, + { + "epoch": 2.1176470588235294, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, + "step": 180 + }, + { + "epoch": 2.3529411764705883, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, + "step": 200 + }, + { + "epoch": 2.3529411764705883, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, + "step": 200 + }, + { + "epoch": 2.588235294117647, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, + "step": 220 + }, + { + "epoch": 2.8235294117647056, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, + "step": 240 + }, + { + "epoch": 3.0588235294117645, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, + "step": 260 + }, + { + "epoch": 3.2941176470588234, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, + "step": 280 + }, + { + "epoch": 3.5294117647058822, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, + "step": 300 + }, + { + "epoch": 3.764705882352941, + "grad_norm": 2.77553391456604, + "learning_rate": 7.1066666666666675e-06, + "loss": 0.2342, + "step": 320 + }, + { + "epoch": 4.0, + "grad_norm": 3.944244861602783, + "learning_rate": 6.8400000000000014e-06, + "loss": 0.1751, + "step": 340 + }, + { + "epoch": 4.235294117647059, + "grad_norm": 6.829845905303955, + "learning_rate": 6.573333333333334e-06, + "loss": 0.1431, + "step": 360 + }, + { + "epoch": 4.470588235294118, + "grad_norm": 3.4126970767974854, + "learning_rate": 6.3066666666666676e-06, + "loss": 0.1468, + "step": 380 + }, + { + "epoch": 4.705882352941177, + "grad_norm": 3.5233240127563477, + "learning_rate": 6.040000000000001e-06, + "loss": 0.1376, + "step": 400 + }, + { + "epoch": 4.705882352941177, + "eval_loss": 0.27411767840385437, + "eval_runtime": 307.4101, + "eval_samples_per_second": 2.215, + "eval_steps_per_second": 0.28, + "eval_wer": 25.511409562892496, + "step": 400 + }, + { + "epoch": 4.9411764705882355, + "grad_norm": 2.9697437286376953, + "learning_rate": 5.7733333333333345e-06, + "loss": 0.1395, + "step": 420 + }, + { + "epoch": 5.176470588235294, + "grad_norm": 3.495859384536743, + "learning_rate": 5.506666666666667e-06, + "loss": 0.1046, + "step": 440 + }, + { + "epoch": 5.411764705882353, + "grad_norm": 4.7592902183532715, + "learning_rate": 5.240000000000001e-06, + "loss": 0.1179, + "step": 460 + }, + { + "epoch": 5.647058823529412, + "grad_norm": 7.579683303833008, + "learning_rate": 4.973333333333334e-06, + "loss": 0.1361, + "step": 480 + }, + { + "epoch": 5.882352941176471, + "grad_norm": 4.949639797210693, + "learning_rate": 4.706666666666667e-06, + "loss": 0.0935, + "step": 500 + }, + { + "epoch": 5.882352941176471, + "eval_loss": 0.2585192918777466, + "eval_runtime": 321.1171, + "eval_samples_per_second": 2.121, + "eval_steps_per_second": 0.268, + "eval_wer": 26.26997450724367, + "step": 500 + }, + { + "epoch": 6.117647058823529, + "grad_norm": 3.536499500274658, + "learning_rate": 4.440000000000001e-06, + "loss": 0.0917, + "step": 520 + }, + { + "epoch": 6.352941176470588, + "grad_norm": 4.393896579742432, + "learning_rate": 4.173333333333334e-06, + "loss": 0.0959, + "step": 540 + }, + { + "epoch": 6.588235294117647, + "grad_norm": 6.232944011688232, + "learning_rate": 3.906666666666667e-06, + "loss": 0.0844, + "step": 560 + }, + { + "epoch": 6.823529411764706, + "grad_norm": 3.1390299797058105, + "learning_rate": 3.6400000000000003e-06, + "loss": 0.0734, + "step": 580 + }, + { + "epoch": 7.0588235294117645, + "grad_norm": 0.599709153175354, + "learning_rate": 3.3733333333333334e-06, + "loss": 0.0575, + "step": 600 + }, + { + "epoch": 7.0588235294117645, + "eval_loss": 0.24722784757614136, + "eval_runtime": 320.2621, + "eval_samples_per_second": 2.126, + "eval_steps_per_second": 0.269, + "eval_wer": 24.187029783000686, + "step": 600 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 5.540839686144e+18, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-600/training_args.bin b/checkpoint-600/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-600/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432 diff --git a/checkpoint-700/config.json b/checkpoint-700/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-700/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-700/generation_config.json b/checkpoint-700/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-700/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-700/model.safetensors b/checkpoint-700/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c45f0aa44a0153b1f094519ac31c7ef4e3f8f3e8 --- /dev/null +++ b/checkpoint-700/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75a07ce2e1473b91dc1d5fb912a14f3698a9be16091de36c0a9e8842dd9cb0df +size 966995080 diff --git a/checkpoint-700/optimizer.pt b/checkpoint-700/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..e1db4d9c31119fcde1d91f0b66748016c2b2d120 --- /dev/null +++ b/checkpoint-700/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:048ee13ee60d7112918d624228122a090f7fe82221fa4d780149142887d4af66 +size 1925064044 diff --git a/checkpoint-700/preprocessor_config.json b/checkpoint-700/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-700/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-700/rng_state.pth b/checkpoint-700/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..8ae68cbb82f90e9b51033caa217949f5572736ff --- /dev/null +++ b/checkpoint-700/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be96f635ab48bd8d0fd9cf141c4430e4f2dd09c9dc899f1e957d08dc2e5a910d +size 14244 diff --git a/checkpoint-700/scaler.pt b/checkpoint-700/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..30665d725d451435d7da81024e92ebd61b653a28 --- /dev/null +++ b/checkpoint-700/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:211adda3e5976d95111c14132862d9365b06bb99c108e89fdd0ad775b46678e9 +size 988 diff --git a/checkpoint-700/scheduler.pt b/checkpoint-700/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..eee9cb4347b24203e07e21ba7c4d4c3cfad52aa1 --- /dev/null +++ b/checkpoint-700/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:adb16e99178a46012c67a4b4d91c2b43733a8bdecab272535bb9ee8ffb1aa83b +size 1064 diff --git a/checkpoint-700/trainer_state.json b/checkpoint-700/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..c11c0a6736f16b3d7122f978fb836891ea3540c5 --- /dev/null +++ b/checkpoint-700/trainer_state.json @@ -0,0 +1,342 @@ +{ + "best_global_step": 700, + "best_metric": 23.397376111422, + "best_model_checkpoint": "./JUDIC/checkpoint-700", + "epoch": 8.235294117647058, + "eval_steps": 100, + "global_step": 700, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + }, + { + "epoch": 1.4117647058823528, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, + "step": 120 + }, + { + "epoch": 1.6470588235294117, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, + "step": 140 + }, + { + "epoch": 1.8823529411764706, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, + "step": 160 + }, + { + "epoch": 2.1176470588235294, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, + "step": 180 + }, + { + "epoch": 2.3529411764705883, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, + "step": 200 + }, + { + "epoch": 2.3529411764705883, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, + "step": 200 + }, + { + "epoch": 2.588235294117647, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, + "step": 220 + }, + { + "epoch": 2.8235294117647056, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, + "step": 240 + }, + { + "epoch": 3.0588235294117645, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, + "step": 260 + }, + { + "epoch": 3.2941176470588234, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, + "step": 280 + }, + { + "epoch": 3.5294117647058822, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, + "step": 300 + }, + { + "epoch": 3.764705882352941, + "grad_norm": 2.77553391456604, + "learning_rate": 7.1066666666666675e-06, + "loss": 0.2342, + "step": 320 + }, + { + "epoch": 4.0, + "grad_norm": 3.944244861602783, + "learning_rate": 6.8400000000000014e-06, + "loss": 0.1751, + "step": 340 + }, + { + "epoch": 4.235294117647059, + "grad_norm": 6.829845905303955, + "learning_rate": 6.573333333333334e-06, + "loss": 0.1431, + "step": 360 + }, + { + "epoch": 4.470588235294118, + "grad_norm": 3.4126970767974854, + "learning_rate": 6.3066666666666676e-06, + "loss": 0.1468, + "step": 380 + }, + { + "epoch": 4.705882352941177, + "grad_norm": 3.5233240127563477, + "learning_rate": 6.040000000000001e-06, + "loss": 0.1376, + "step": 400 + }, + { + "epoch": 4.705882352941177, + "eval_loss": 0.27411767840385437, + "eval_runtime": 307.4101, + "eval_samples_per_second": 2.215, + "eval_steps_per_second": 0.28, + "eval_wer": 25.511409562892496, + "step": 400 + }, + { + "epoch": 4.9411764705882355, + "grad_norm": 2.9697437286376953, + "learning_rate": 5.7733333333333345e-06, + "loss": 0.1395, + "step": 420 + }, + { + "epoch": 5.176470588235294, + "grad_norm": 3.495859384536743, + "learning_rate": 5.506666666666667e-06, + "loss": 0.1046, + "step": 440 + }, + { + "epoch": 5.411764705882353, + "grad_norm": 4.7592902183532715, + "learning_rate": 5.240000000000001e-06, + "loss": 0.1179, + "step": 460 + }, + { + "epoch": 5.647058823529412, + "grad_norm": 7.579683303833008, + "learning_rate": 4.973333333333334e-06, + "loss": 0.1361, + "step": 480 + }, + { + "epoch": 5.882352941176471, + "grad_norm": 4.949639797210693, + "learning_rate": 4.706666666666667e-06, + "loss": 0.0935, + "step": 500 + }, + { + "epoch": 5.882352941176471, + "eval_loss": 0.2585192918777466, + "eval_runtime": 321.1171, + "eval_samples_per_second": 2.121, + "eval_steps_per_second": 0.268, + "eval_wer": 26.26997450724367, + "step": 500 + }, + { + "epoch": 6.117647058823529, + "grad_norm": 3.536499500274658, + "learning_rate": 4.440000000000001e-06, + "loss": 0.0917, + "step": 520 + }, + { + "epoch": 6.352941176470588, + "grad_norm": 4.393896579742432, + "learning_rate": 4.173333333333334e-06, + "loss": 0.0959, + "step": 540 + }, + { + "epoch": 6.588235294117647, + "grad_norm": 6.232944011688232, + "learning_rate": 3.906666666666667e-06, + "loss": 0.0844, + "step": 560 + }, + { + "epoch": 6.823529411764706, + "grad_norm": 3.1390299797058105, + "learning_rate": 3.6400000000000003e-06, + "loss": 0.0734, + "step": 580 + }, + { + "epoch": 7.0588235294117645, + "grad_norm": 0.599709153175354, + "learning_rate": 3.3733333333333334e-06, + "loss": 0.0575, + "step": 600 + }, + { + "epoch": 7.0588235294117645, + "eval_loss": 0.24722784757614136, + "eval_runtime": 320.2621, + "eval_samples_per_second": 2.126, + "eval_steps_per_second": 0.269, + "eval_wer": 24.187029783000686, + "step": 600 + }, + { + "epoch": 7.294117647058823, + "grad_norm": 7.411947727203369, + "learning_rate": 3.106666666666667e-06, + "loss": 0.0598, + "step": 620 + }, + { + "epoch": 7.529411764705882, + "grad_norm": 0.39204472303390503, + "learning_rate": 2.84e-06, + "loss": 0.045, + "step": 640 + }, + { + "epoch": 7.764705882352941, + "grad_norm": 0.8188313245773315, + "learning_rate": 2.573333333333334e-06, + "loss": 0.0598, + "step": 660 + }, + { + "epoch": 8.0, + "grad_norm": 2.0588011741638184, + "learning_rate": 2.306666666666667e-06, + "loss": 0.0663, + "step": 680 + }, + { + "epoch": 8.235294117647058, + "grad_norm": 0.8121541142463684, + "learning_rate": 2.04e-06, + "loss": 0.0424, + "step": 700 + }, + { + "epoch": 8.235294117647058, + "eval_loss": 0.24150076508522034, + "eval_runtime": 319.2037, + "eval_samples_per_second": 2.133, + "eval_steps_per_second": 0.269, + "eval_wer": 23.397376111422, + "step": 700 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.464312967168e+18, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-700/training_args.bin b/checkpoint-700/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-700/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432 diff --git a/checkpoint-800/config.json b/checkpoint-800/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-800/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-800/generation_config.json b/checkpoint-800/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-800/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-800/model.safetensors b/checkpoint-800/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4e2c0a5c4a3be78617965cb145953d8087e18a26 --- /dev/null +++ b/checkpoint-800/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fcdf91cb2efe88f38390d0d407e828c0d192a474dfc15fe3a2b5358a7f2967b +size 966995080 diff --git a/checkpoint-800/optimizer.pt b/checkpoint-800/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..2c4808b5a80e7f460649372a4cdccf9c820c6f01 --- /dev/null +++ b/checkpoint-800/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:225146d46e40502786d82584c88ff81c56e32d0292f419be70cab4a1414ea72a +size 1925064044 diff --git a/checkpoint-800/preprocessor_config.json b/checkpoint-800/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-800/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-800/rng_state.pth b/checkpoint-800/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..a631426eaf3aec45bb137851e3189c5f3abf0b7f --- /dev/null +++ b/checkpoint-800/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4865bd208465b21416bcee585790a0f6d4ff26461be9f538bcc3b8df73f51bc1 +size 14244 diff --git a/checkpoint-800/scaler.pt b/checkpoint-800/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..1db85c6b10709bdf9c1a4b519cacd4bb788a048f --- /dev/null +++ b/checkpoint-800/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bef6eaad775e6c5db9d59a18290abbf38e5801de022da76bb8150ebd7d9dc841 +size 988 diff --git a/checkpoint-800/scheduler.pt b/checkpoint-800/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..3594db3c39cd7c2a49b5f9aa70c06c35caf02cf6 --- /dev/null +++ b/checkpoint-800/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:180ba15249ffbfeafef27a842c80652611e6770d582ad4bd65efde638c7011e3 +size 1064 diff --git a/checkpoint-800/trainer_state.json b/checkpoint-800/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..b0fd5ec6202faa7fc2ee97fc274d4a629efcf538 --- /dev/null +++ b/checkpoint-800/trainer_state.json @@ -0,0 +1,386 @@ +{ + "best_global_step": 800, + "best_metric": 21.18385873282348, + "best_model_checkpoint": "./JUDIC/checkpoint-800", + "epoch": 9.411764705882353, + "eval_steps": 100, + "global_step": 800, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + }, + { + "epoch": 1.4117647058823528, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, + "step": 120 + }, + { + "epoch": 1.6470588235294117, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, + "step": 140 + }, + { + "epoch": 1.8823529411764706, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, + "step": 160 + }, + { + "epoch": 2.1176470588235294, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, + "step": 180 + }, + { + "epoch": 2.3529411764705883, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, + "step": 200 + }, + { + "epoch": 2.3529411764705883, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, + "step": 200 + }, + { + "epoch": 2.588235294117647, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, + "step": 220 + }, + { + "epoch": 2.8235294117647056, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, + "step": 240 + }, + { + "epoch": 3.0588235294117645, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, + "step": 260 + }, + { + "epoch": 3.2941176470588234, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, + "step": 280 + }, + { + "epoch": 3.5294117647058822, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, + "step": 300 + }, + { + "epoch": 3.764705882352941, + "grad_norm": 2.77553391456604, + "learning_rate": 7.1066666666666675e-06, + "loss": 0.2342, + "step": 320 + }, + { + "epoch": 4.0, + "grad_norm": 3.944244861602783, + "learning_rate": 6.8400000000000014e-06, + "loss": 0.1751, + "step": 340 + }, + { + "epoch": 4.235294117647059, + "grad_norm": 6.829845905303955, + "learning_rate": 6.573333333333334e-06, + "loss": 0.1431, + "step": 360 + }, + { + "epoch": 4.470588235294118, + "grad_norm": 3.4126970767974854, + "learning_rate": 6.3066666666666676e-06, + "loss": 0.1468, + "step": 380 + }, + { + "epoch": 4.705882352941177, + "grad_norm": 3.5233240127563477, + "learning_rate": 6.040000000000001e-06, + "loss": 0.1376, + "step": 400 + }, + { + "epoch": 4.705882352941177, + "eval_loss": 0.27411767840385437, + "eval_runtime": 307.4101, + "eval_samples_per_second": 2.215, + "eval_steps_per_second": 0.28, + "eval_wer": 25.511409562892496, + "step": 400 + }, + { + "epoch": 4.9411764705882355, + "grad_norm": 2.9697437286376953, + "learning_rate": 5.7733333333333345e-06, + "loss": 0.1395, + "step": 420 + }, + { + "epoch": 5.176470588235294, + "grad_norm": 3.495859384536743, + "learning_rate": 5.506666666666667e-06, + "loss": 0.1046, + "step": 440 + }, + { + "epoch": 5.411764705882353, + "grad_norm": 4.7592902183532715, + "learning_rate": 5.240000000000001e-06, + "loss": 0.1179, + "step": 460 + }, + { + "epoch": 5.647058823529412, + "grad_norm": 7.579683303833008, + "learning_rate": 4.973333333333334e-06, + "loss": 0.1361, + "step": 480 + }, + { + "epoch": 5.882352941176471, + "grad_norm": 4.949639797210693, + "learning_rate": 4.706666666666667e-06, + "loss": 0.0935, + "step": 500 + }, + { + "epoch": 5.882352941176471, + "eval_loss": 0.2585192918777466, + "eval_runtime": 321.1171, + "eval_samples_per_second": 2.121, + "eval_steps_per_second": 0.268, + "eval_wer": 26.26997450724367, + "step": 500 + }, + { + "epoch": 6.117647058823529, + "grad_norm": 3.536499500274658, + "learning_rate": 4.440000000000001e-06, + "loss": 0.0917, + "step": 520 + }, + { + "epoch": 6.352941176470588, + "grad_norm": 4.393896579742432, + "learning_rate": 4.173333333333334e-06, + "loss": 0.0959, + "step": 540 + }, + { + "epoch": 6.588235294117647, + "grad_norm": 6.232944011688232, + "learning_rate": 3.906666666666667e-06, + "loss": 0.0844, + "step": 560 + }, + { + "epoch": 6.823529411764706, + "grad_norm": 3.1390299797058105, + "learning_rate": 3.6400000000000003e-06, + "loss": 0.0734, + "step": 580 + }, + { + "epoch": 7.0588235294117645, + "grad_norm": 0.599709153175354, + "learning_rate": 3.3733333333333334e-06, + "loss": 0.0575, + "step": 600 + }, + { + "epoch": 7.0588235294117645, + "eval_loss": 0.24722784757614136, + "eval_runtime": 320.2621, + "eval_samples_per_second": 2.126, + "eval_steps_per_second": 0.269, + "eval_wer": 24.187029783000686, + "step": 600 + }, + { + "epoch": 7.294117647058823, + "grad_norm": 7.411947727203369, + "learning_rate": 3.106666666666667e-06, + "loss": 0.0598, + "step": 620 + }, + { + "epoch": 7.529411764705882, + "grad_norm": 0.39204472303390503, + "learning_rate": 2.84e-06, + "loss": 0.045, + "step": 640 + }, + { + "epoch": 7.764705882352941, + "grad_norm": 0.8188313245773315, + "learning_rate": 2.573333333333334e-06, + "loss": 0.0598, + "step": 660 + }, + { + "epoch": 8.0, + "grad_norm": 2.0588011741638184, + "learning_rate": 2.306666666666667e-06, + "loss": 0.0663, + "step": 680 + }, + { + "epoch": 8.235294117647058, + "grad_norm": 0.8121541142463684, + "learning_rate": 2.04e-06, + "loss": 0.0424, + "step": 700 + }, + { + "epoch": 8.235294117647058, + "eval_loss": 0.24150076508522034, + "eval_runtime": 319.2037, + "eval_samples_per_second": 2.133, + "eval_steps_per_second": 0.269, + "eval_wer": 23.397376111422, + "step": 700 + }, + { + "epoch": 8.470588235294118, + "grad_norm": 1.8647712469100952, + "learning_rate": 1.7733333333333336e-06, + "loss": 0.0466, + "step": 720 + }, + { + "epoch": 8.705882352941176, + "grad_norm": 0.49962764978408813, + "learning_rate": 1.506666666666667e-06, + "loss": 0.0532, + "step": 740 + }, + { + "epoch": 8.941176470588236, + "grad_norm": 1.0500967502593994, + "learning_rate": 1.2400000000000002e-06, + "loss": 0.0406, + "step": 760 + }, + { + "epoch": 9.176470588235293, + "grad_norm": 2.893846035003662, + "learning_rate": 9.733333333333333e-07, + "loss": 0.0326, + "step": 780 + }, + { + "epoch": 9.411764705882353, + "grad_norm": 1.594116449356079, + "learning_rate": 7.066666666666667e-07, + "loss": 0.0423, + "step": 800 + }, + { + "epoch": 9.411764705882353, + "eval_loss": 0.23845414817333221, + "eval_runtime": 313.826, + "eval_samples_per_second": 2.17, + "eval_steps_per_second": 0.274, + "eval_wer": 21.18385873282348, + "step": 800 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 7.387786248192e+18, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-800/training_args.bin b/checkpoint-800/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-800/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432 diff --git a/checkpoint-850/config.json b/checkpoint-850/config.json new file mode 100644 index 0000000000000000000000000000000000000000..337642e18341c758364c32c6b6159a090615329b --- /dev/null +++ b/checkpoint-850/config.json @@ -0,0 +1,47 @@ +{ + "activation_dropout": 0.0, + "activation_function": "gelu", + "apply_spec_augment": false, + "architectures": [ + "WhisperForConditionalGeneration" + ], + "attention_dropout": 0.0, + "begin_suppress_tokens": null, + "bos_token_id": 50257, + "classifier_proj_size": 256, + "d_model": 768, + "decoder_attention_heads": 12, + "decoder_ffn_dim": 3072, + "decoder_layerdrop": 0.0, + "decoder_layers": 12, + "decoder_start_token_id": 50258, + "dropout": 0.0, + "encoder_attention_heads": 12, + "encoder_ffn_dim": 3072, + "encoder_layerdrop": 0.0, + "encoder_layers": 12, + "eos_token_id": 50257, + "forced_decoder_ids": null, + "init_std": 0.02, + "is_encoder_decoder": true, + "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, + "max_length": null, + "max_source_positions": 1500, + "max_target_positions": 448, + "median_filter_width": 7, + "model_type": "whisper", + "num_hidden_layers": 12, + "num_mel_bins": 80, + "pad_token_id": 50257, + "scale_embedding": false, + "torch_dtype": "float32", + "transformers_version": "4.51.3", + "use_cache": false, + "use_weighted_layer_sum": false, + "vocab_size": 51865 +} diff --git a/checkpoint-850/generation_config.json b/checkpoint-850/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ec4263613bc098f8eb01034b73fbe16fb43eba5c --- /dev/null +++ b/checkpoint-850/generation_config.json @@ -0,0 +1,175 @@ +{ + "alignment_heads": [ + [ + 5, + 3 + ], + [ + 5, + 9 + ], + [ + 8, + 0 + ], + [ + 8, + 4 + ], + [ + 8, + 7 + ], + [ + 8, + 8 + ], + [ + 9, + 0 + ], + [ + 9, + 7 + ], + [ + 9, + 9 + ], + [ + 10, + 5 + ] + ], + "begin_suppress_tokens": [ + 220, + 50257 + ], + "bos_token_id": 50257, + "decoder_start_token_id": 50258, + "eos_token_id": 50257, + "forced_decoder_ids": [ + [ + 1, + null + ], + [ + 2, + 50359 + ] + ], + "is_multilingual": true, + "lang_to_id": { + "<|af|>": 50327, + "<|am|>": 50334, + "<|ar|>": 50272, + "<|as|>": 50350, + "<|az|>": 50304, + "<|ba|>": 50355, + "<|be|>": 50330, + "<|bg|>": 50292, + "<|bn|>": 50302, + "<|bo|>": 50347, + "<|br|>": 50309, + "<|bs|>": 50315, + "<|ca|>": 50270, + "<|cs|>": 50283, + "<|cy|>": 50297, + "<|da|>": 50285, + "<|de|>": 50261, + "<|el|>": 50281, + "<|en|>": 50259, + "<|es|>": 50262, + "<|et|>": 50307, + "<|eu|>": 50310, + "<|fa|>": 50300, + "<|fi|>": 50277, + "<|fo|>": 50338, + "<|fr|>": 50265, + "<|gl|>": 50319, + "<|gu|>": 50333, + "<|haw|>": 50352, + "<|ha|>": 50354, + "<|he|>": 50279, + "<|hi|>": 50276, + "<|hr|>": 50291, + "<|ht|>": 50339, + "<|hu|>": 50286, + "<|hy|>": 50312, + "<|id|>": 50275, + "<|is|>": 50311, + "<|it|>": 50274, + "<|ja|>": 50266, + "<|jw|>": 50356, + "<|ka|>": 50329, + "<|kk|>": 50316, + "<|km|>": 50323, + "<|kn|>": 50306, + "<|ko|>": 50264, + "<|la|>": 50294, + "<|lb|>": 50345, + "<|ln|>": 50353, + "<|lo|>": 50336, + "<|lt|>": 50293, + "<|lv|>": 50301, + "<|mg|>": 50349, + "<|mi|>": 50295, + "<|mk|>": 50308, + "<|ml|>": 50296, + "<|mn|>": 50314, + "<|mr|>": 50320, + "<|ms|>": 50282, + "<|mt|>": 50343, + "<|my|>": 50346, + "<|ne|>": 50313, + "<|nl|>": 50271, + "<|nn|>": 50342, + "<|no|>": 50288, + "<|oc|>": 50328, + "<|pa|>": 50321, + "<|pl|>": 50269, + "<|ps|>": 50340, + "<|pt|>": 50267, + "<|ro|>": 50284, + "<|ru|>": 50263, + "<|sa|>": 50344, + "<|sd|>": 50332, + "<|si|>": 50322, + "<|sk|>": 50298, + "<|sl|>": 50305, + "<|sn|>": 50324, + "<|so|>": 50326, + "<|sq|>": 50317, + "<|sr|>": 50303, + "<|su|>": 50357, + "<|sv|>": 50273, + "<|sw|>": 50318, + "<|ta|>": 50287, + "<|te|>": 50299, + "<|tg|>": 50331, + "<|th|>": 50289, + "<|tk|>": 50341, + "<|tl|>": 50348, + "<|tr|>": 50268, + "<|tt|>": 50351, + "<|uk|>": 50280, + "<|ur|>": 50290, + "<|uz|>": 50337, + "<|vi|>": 50278, + "<|yi|>": 50335, + "<|yo|>": 50325, + "<|zh|>": 50260 + }, + "max_initial_timestamp_index": 50, + "max_length": 448, + "no_timestamps_token_id": 50363, + "pad_token_id": 50257, + "prev_sot_token_id": 50361, + "return_timestamps": false, + "suppress_tokens": [], + "task_to_id": { + "transcribe": 50359, + "translate": 50358 + }, + "transformers_version": "4.51.3" +} diff --git a/checkpoint-850/model.safetensors b/checkpoint-850/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..79c98e36d0dea03f90320a2ffcc654850e6f550a --- /dev/null +++ b/checkpoint-850/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ea8d6049c09849e2d291bd293939312078ec95626ac1f4897ce37dfe4fd8caf +size 966995080 diff --git a/checkpoint-850/optimizer.pt b/checkpoint-850/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..287a84346648eabf71a3524531635b589a2d4102 --- /dev/null +++ b/checkpoint-850/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a891ab244072775b903d78ed33dc073e1deec5bf259c29e7f5fecd2417ca58f +size 1925064044 diff --git a/checkpoint-850/preprocessor_config.json b/checkpoint-850/preprocessor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d7c5b1dfd466c815664b16139e8f605620d22e4e --- /dev/null +++ b/checkpoint-850/preprocessor_config.json @@ -0,0 +1,15 @@ +{ + "chunk_length": 30, + "dither": 0.0, + "feature_extractor_type": "WhisperFeatureExtractor", + "feature_size": 80, + "hop_length": 160, + "n_fft": 400, + "n_samples": 480000, + "nb_max_frames": 3000, + "padding_side": "right", + "padding_value": 0.0, + "processor_class": "WhisperProcessor", + "return_attention_mask": false, + "sampling_rate": 16000 +} diff --git a/checkpoint-850/rng_state.pth b/checkpoint-850/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..0f1a5467f7505eeb7fd3a7d59fff73fcb07e4c27 --- /dev/null +++ b/checkpoint-850/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2c45272ffded18a9683bfa9aee24d5384a1c2b9434b81ecb36df37cff68f1105 +size 14244 diff --git a/checkpoint-850/scaler.pt b/checkpoint-850/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..1fa3f31829397c77936dfca3cc851aac645db28a --- /dev/null +++ b/checkpoint-850/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5f376192cd567dd4223c334d641bccc0a038410131bf0f61d6655762af402954 +size 988 diff --git a/checkpoint-850/scheduler.pt b/checkpoint-850/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..6273d20759a7e9cfaa41cb90c916f25ffb5ed6e6 --- /dev/null +++ b/checkpoint-850/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:addb94f000bf4ff6afcfd865969db46b21c56041e9a4a9baef556c46b1c07096 +size 1064 diff --git a/checkpoint-850/trainer_state.json b/checkpoint-850/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8a610820e49c7274be3fe806d0ebe140e4958708 --- /dev/null +++ b/checkpoint-850/trainer_state.json @@ -0,0 +1,400 @@ +{ + "best_global_step": 800, + "best_metric": 21.18385873282348, + "best_model_checkpoint": "./JUDIC/checkpoint-800", + "epoch": 10.0, + "eval_steps": 100, + "global_step": 850, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.23529411764705882, + "grad_norm": 4.239101886749268, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.4078, + "step": 20 + }, + { + "epoch": 0.47058823529411764, + "grad_norm": 2.235217809677124, + "learning_rate": 3.900000000000001e-06, + "loss": 0.4217, + "step": 40 + }, + { + "epoch": 0.7058823529411765, + "grad_norm": 3.711195707321167, + "learning_rate": 5.9e-06, + "loss": 0.4252, + "step": 60 + }, + { + "epoch": 0.9411764705882353, + "grad_norm": 4.5320820808410645, + "learning_rate": 7.9e-06, + "loss": 0.3812, + "step": 80 + }, + { + "epoch": 1.1764705882352942, + "grad_norm": 3.9703948497772217, + "learning_rate": 9.9e-06, + "loss": 0.4223, + "step": 100 + }, + { + "epoch": 1.1764705882352942, + "eval_loss": 0.40899839997291565, + "eval_runtime": 323.6745, + "eval_samples_per_second": 2.104, + "eval_steps_per_second": 0.266, + "eval_wer": 36.6100851831126, + "step": 100 + }, + { + "epoch": 1.4117647058823528, + "grad_norm": 3.8265273571014404, + "learning_rate": 9.746666666666668e-06, + "loss": 0.2872, + "step": 120 + }, + { + "epoch": 1.6470588235294117, + "grad_norm": 4.8625168800354, + "learning_rate": 9.48e-06, + "loss": 0.3207, + "step": 140 + }, + { + "epoch": 1.8823529411764706, + "grad_norm": 4.235057830810547, + "learning_rate": 9.213333333333334e-06, + "loss": 0.3585, + "step": 160 + }, + { + "epoch": 2.1176470588235294, + "grad_norm": 2.1225504875183105, + "learning_rate": 8.973333333333334e-06, + "loss": 0.2658, + "step": 180 + }, + { + "epoch": 2.3529411764705883, + "grad_norm": 8.930822372436523, + "learning_rate": 8.706666666666667e-06, + "loss": 0.2576, + "step": 200 + }, + { + "epoch": 2.3529411764705883, + "eval_loss": 0.35160648822784424, + "eval_runtime": 296.116, + "eval_samples_per_second": 2.3, + "eval_steps_per_second": 0.29, + "eval_wer": 29.938444320089534, + "step": 200 + }, + { + "epoch": 2.588235294117647, + "grad_norm": 1.942764163017273, + "learning_rate": 8.44e-06, + "loss": 0.283, + "step": 220 + }, + { + "epoch": 2.8235294117647056, + "grad_norm": 3.2245473861694336, + "learning_rate": 8.173333333333334e-06, + "loss": 0.2443, + "step": 240 + }, + { + "epoch": 3.0588235294117645, + "grad_norm": 3.0929267406463623, + "learning_rate": 7.906666666666667e-06, + "loss": 0.2489, + "step": 260 + }, + { + "epoch": 3.2941176470588234, + "grad_norm": 3.477473497390747, + "learning_rate": 7.640000000000001e-06, + "loss": 0.2006, + "step": 280 + }, + { + "epoch": 3.5294117647058822, + "grad_norm": 3.163804531097412, + "learning_rate": 7.373333333333334e-06, + "loss": 0.1837, + "step": 300 + }, + { + "epoch": 3.5294117647058822, + "eval_loss": 0.30386850237846375, + "eval_runtime": 329.1958, + "eval_samples_per_second": 2.069, + "eval_steps_per_second": 0.261, + "eval_wer": 31.555058135919918, + "step": 300 + }, + { + "epoch": 3.764705882352941, + "grad_norm": 2.77553391456604, + "learning_rate": 7.1066666666666675e-06, + "loss": 0.2342, + "step": 320 + }, + { + "epoch": 4.0, + "grad_norm": 3.944244861602783, + "learning_rate": 6.8400000000000014e-06, + "loss": 0.1751, + "step": 340 + }, + { + "epoch": 4.235294117647059, + "grad_norm": 6.829845905303955, + "learning_rate": 6.573333333333334e-06, + "loss": 0.1431, + "step": 360 + }, + { + "epoch": 4.470588235294118, + "grad_norm": 3.4126970767974854, + "learning_rate": 6.3066666666666676e-06, + "loss": 0.1468, + "step": 380 + }, + { + "epoch": 4.705882352941177, + "grad_norm": 3.5233240127563477, + "learning_rate": 6.040000000000001e-06, + "loss": 0.1376, + "step": 400 + }, + { + "epoch": 4.705882352941177, + "eval_loss": 0.27411767840385437, + "eval_runtime": 307.4101, + "eval_samples_per_second": 2.215, + "eval_steps_per_second": 0.28, + "eval_wer": 25.511409562892496, + "step": 400 + }, + { + "epoch": 4.9411764705882355, + "grad_norm": 2.9697437286376953, + "learning_rate": 5.7733333333333345e-06, + "loss": 0.1395, + "step": 420 + }, + { + "epoch": 5.176470588235294, + "grad_norm": 3.495859384536743, + "learning_rate": 5.506666666666667e-06, + "loss": 0.1046, + "step": 440 + }, + { + "epoch": 5.411764705882353, + "grad_norm": 4.7592902183532715, + "learning_rate": 5.240000000000001e-06, + "loss": 0.1179, + "step": 460 + }, + { + "epoch": 5.647058823529412, + "grad_norm": 7.579683303833008, + "learning_rate": 4.973333333333334e-06, + "loss": 0.1361, + "step": 480 + }, + { + "epoch": 5.882352941176471, + "grad_norm": 4.949639797210693, + "learning_rate": 4.706666666666667e-06, + "loss": 0.0935, + "step": 500 + }, + { + "epoch": 5.882352941176471, + "eval_loss": 0.2585192918777466, + "eval_runtime": 321.1171, + "eval_samples_per_second": 2.121, + "eval_steps_per_second": 0.268, + "eval_wer": 26.26997450724367, + "step": 500 + }, + { + "epoch": 6.117647058823529, + "grad_norm": 3.536499500274658, + "learning_rate": 4.440000000000001e-06, + "loss": 0.0917, + "step": 520 + }, + { + "epoch": 6.352941176470588, + "grad_norm": 4.393896579742432, + "learning_rate": 4.173333333333334e-06, + "loss": 0.0959, + "step": 540 + }, + { + "epoch": 6.588235294117647, + "grad_norm": 6.232944011688232, + "learning_rate": 3.906666666666667e-06, + "loss": 0.0844, + "step": 560 + }, + { + "epoch": 6.823529411764706, + "grad_norm": 3.1390299797058105, + "learning_rate": 3.6400000000000003e-06, + "loss": 0.0734, + "step": 580 + }, + { + "epoch": 7.0588235294117645, + "grad_norm": 0.599709153175354, + "learning_rate": 3.3733333333333334e-06, + "loss": 0.0575, + "step": 600 + }, + { + "epoch": 7.0588235294117645, + "eval_loss": 0.24722784757614136, + "eval_runtime": 320.2621, + "eval_samples_per_second": 2.126, + "eval_steps_per_second": 0.269, + "eval_wer": 24.187029783000686, + "step": 600 + }, + { + "epoch": 7.294117647058823, + "grad_norm": 7.411947727203369, + "learning_rate": 3.106666666666667e-06, + "loss": 0.0598, + "step": 620 + }, + { + "epoch": 7.529411764705882, + "grad_norm": 0.39204472303390503, + "learning_rate": 2.84e-06, + "loss": 0.045, + "step": 640 + }, + { + "epoch": 7.764705882352941, + "grad_norm": 0.8188313245773315, + "learning_rate": 2.573333333333334e-06, + "loss": 0.0598, + "step": 660 + }, + { + "epoch": 8.0, + "grad_norm": 2.0588011741638184, + "learning_rate": 2.306666666666667e-06, + "loss": 0.0663, + "step": 680 + }, + { + "epoch": 8.235294117647058, + "grad_norm": 0.8121541142463684, + "learning_rate": 2.04e-06, + "loss": 0.0424, + "step": 700 + }, + { + "epoch": 8.235294117647058, + "eval_loss": 0.24150076508522034, + "eval_runtime": 319.2037, + "eval_samples_per_second": 2.133, + "eval_steps_per_second": 0.269, + "eval_wer": 23.397376111422, + "step": 700 + }, + { + "epoch": 8.470588235294118, + "grad_norm": 1.8647712469100952, + "learning_rate": 1.7733333333333336e-06, + "loss": 0.0466, + "step": 720 + }, + { + "epoch": 8.705882352941176, + "grad_norm": 0.49962764978408813, + "learning_rate": 1.506666666666667e-06, + "loss": 0.0532, + "step": 740 + }, + { + "epoch": 8.941176470588236, + "grad_norm": 1.0500967502593994, + "learning_rate": 1.2400000000000002e-06, + "loss": 0.0406, + "step": 760 + }, + { + "epoch": 9.176470588235293, + "grad_norm": 2.893846035003662, + "learning_rate": 9.733333333333333e-07, + "loss": 0.0326, + "step": 780 + }, + { + "epoch": 9.411764705882353, + "grad_norm": 1.594116449356079, + "learning_rate": 7.066666666666667e-07, + "loss": 0.0423, + "step": 800 + }, + { + "epoch": 9.411764705882353, + "eval_loss": 0.23845414817333221, + "eval_runtime": 313.826, + "eval_samples_per_second": 2.17, + "eval_steps_per_second": 0.274, + "eval_wer": 21.18385873282348, + "step": 800 + }, + { + "epoch": 9.647058823529411, + "grad_norm": 0.34508946537971497, + "learning_rate": 4.4e-07, + "loss": 0.0162, + "step": 820 + }, + { + "epoch": 9.882352941176471, + "grad_norm": 0.9907822012901306, + "learning_rate": 1.7333333333333335e-07, + "loss": 0.029, + "step": 840 + } + ], + "logging_steps": 20, + "max_steps": 850, + "num_input_tokens_seen": 0, + "num_train_epochs": 10, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 7.849522888704e+18, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-850/training_args.bin b/checkpoint-850/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..dda69251eaface4172ea0a644fb661871a9152e7 --- /dev/null +++ b/checkpoint-850/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a582364a1defc1cdd820215263292cba4d325748959fa5eb84ed58c1bcccb33d +size 5432