End of training

README.md

@@ -0,0 +1,81 @@
+---
+license: apache-2.0
+base_model: facebook/hubert-large-ls960-ft
+tags:
+- generated_from_trainer
+metrics:
+- wer
+model-index:
+- name: hubert-large-ls960-ft-V2-5
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# hubert-large-ls960-ft-V2-5
+
+This model is a fine-tuned version of [facebook/hubert-large-ls960-ft](https://huggingface.co/facebook/hubert-large-ls960-ft) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5760
+- Wer: 0.1085
+- Per: 0.0892
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 20
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    | Per    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
+| 19.707        | 1.0   | 82   | 3.5254          | 1.0    | 1.0    |
+| 3.4906        | 2.0   | 164  | 3.2483          | 1.0    | 1.0    |
+| 3.233         | 3.0   | 246  | 3.1368          | 1.0    | 1.0    |
+| 3.0468        | 4.0   | 328  | 2.9600          | 1.0    | 1.0    |
+| 2.6751        | 5.0   | 410  | 2.3348          | 1.0    | 1.0    |
+| 2.0881        | 6.0   | 492  | 1.7351          | 0.8568 | 0.8726 |
+| 1.4875        | 7.0   | 574  | 1.2264          | 0.6059 | 0.6134 |
+| 1.0922        | 8.0   | 656  | 0.9666          | 0.4068 | 0.3972 |
+| 0.8148        | 9.0   | 738  | 0.7746          | 0.3249 | 0.3138 |
+| 0.6332        | 10.0  | 820  | 0.6755          | 0.2477 | 0.2313 |
+| 0.4797        | 11.0  | 902  | 0.6262          | 0.1612 | 0.1410 |
+| 0.3807        | 12.0  | 984  | 0.5765          | 0.1384 | 0.1172 |
+| 0.3195        | 13.0  | 1066 | 0.5666          | 0.1191 | 0.0992 |
+| 0.2526        | 14.0  | 1148 | 0.5759          | 0.1165 | 0.0970 |
+| 0.2417        | 15.0  | 1230 | 0.5460          | 0.1138 | 0.0946 |
+| 0.2072        | 16.0  | 1312 | 0.5551          | 0.1095 | 0.0912 |
+| 0.1881        | 17.0  | 1394 | 0.5745          | 0.1102 | 0.0917 |
+| 0.1888        | 18.0  | 1476 | 0.5731          | 0.1094 | 0.0907 |
+| 0.202         | 19.0  | 1558 | 0.5774          | 0.1081 | 0.0893 |
+| 0.1813        | 20.0  | 1640 | 0.5760          | 0.1085 | 0.0892 |
+
+
+### Framework versions
+
+- Transformers 4.35.2
+- Pytorch 2.1.0+cu121
+- Datasets 2.16.1
+- Tokenizers 0.15.0

config.json

@@ -0,0 +1,76 @@
+{
+  "_name_or_path": "facebook/hubert-large-ls960-ft",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "HubertForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.1,
+  "feat_proj_layer_norm": true,
+  "final_dropout": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "model_type": "hubert",
+  "num_attention_heads": 16,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "pad_token_id": 37,
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 40
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d657a31c21aa76a193d8697fbc3e8b71150a4d2ceacf3057fe8acb8c781b616
+size 1261970648

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2Processor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

trainer_state.json

@@ -0,0 +1,348 @@
+{
+  "best_metric": 0.5459895730018616,
+  "best_model_checkpoint": "nrshoudi/hubert-large-ls960-ft-V2-5/checkpoint-1230",
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 1640,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 5e-05,
+      "loss": 19.707,
+      "step": 82
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.5254127979278564,
+      "eval_per": 1.0,
+      "eval_runtime": 116.6109,
+      "eval_samples_per_second": 6.972,
+      "eval_steps_per_second": 3.49,
+      "eval_wer": 1.0,
+      "step": 82
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0001,
+      "loss": 3.4906,
+      "step": 164
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.2483434677124023,
+      "eval_per": 1.0,
+      "eval_runtime": 115.6174,
+      "eval_samples_per_second": 7.032,
+      "eval_steps_per_second": 3.52,
+      "eval_wer": 1.0,
+      "step": 164
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 9.444444444444444e-05,
+      "loss": 3.233,
+      "step": 246
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.1367974281311035,
+      "eval_per": 1.0,
+      "eval_runtime": 116.034,
+      "eval_samples_per_second": 7.007,
+      "eval_steps_per_second": 3.508,
+      "eval_wer": 1.0,
+      "step": 246
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 8.888888888888889e-05,
+      "loss": 3.0468,
+      "step": 328
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.960031509399414,
+      "eval_per": 1.0,
+      "eval_runtime": 116.4247,
+      "eval_samples_per_second": 6.983,
+      "eval_steps_per_second": 3.496,
+      "eval_wer": 1.0,
+      "step": 328
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 8.333333333333334e-05,
+      "loss": 2.6751,
+      "step": 410
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.334815263748169,
+      "eval_per": 1.0,
+      "eval_runtime": 116.7509,
+      "eval_samples_per_second": 6.964,
+      "eval_steps_per_second": 3.486,
+      "eval_wer": 1.0,
+      "step": 410
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 7.777777777777778e-05,
+      "loss": 2.0881,
+      "step": 492
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.7351281642913818,
+      "eval_per": 0.8726326203712428,
+      "eval_runtime": 117.1554,
+      "eval_samples_per_second": 6.94,
+      "eval_steps_per_second": 3.474,
+      "eval_wer": 0.8568428911390042,
+      "step": 492
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 7.222222222222222e-05,
+      "loss": 1.4875,
+      "step": 574
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.22636878490448,
+      "eval_per": 0.613422218034486,
+      "eval_runtime": 116.8569,
+      "eval_samples_per_second": 6.957,
+      "eval_steps_per_second": 3.483,
+      "eval_wer": 0.6058513368484905,
+      "step": 574
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 1.0922,
+      "step": 656
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.9666064381599426,
+      "eval_per": 0.3971779892584566,
+      "eval_runtime": 117.3764,
+      "eval_samples_per_second": 6.926,
+      "eval_steps_per_second": 3.467,
+      "eval_wer": 0.4067938966917083,
+      "step": 656
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 6.111111111111112e-05,
+      "loss": 0.8148,
+      "step": 738
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.7746477127075195,
+      "eval_per": 0.31381324790351456,
+      "eval_runtime": 117.0529,
+      "eval_samples_per_second": 6.946,
+      "eval_steps_per_second": 3.477,
+      "eval_wer": 0.32485651626149037,
+      "step": 738
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 5.555555555555556e-05,
+      "loss": 0.6332,
+      "step": 820
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.6754913330078125,
+      "eval_per": 0.23132007914821445,
+      "eval_runtime": 117.3105,
+      "eval_samples_per_second": 6.93,
+      "eval_steps_per_second": 3.469,
+      "eval_wer": 0.24767859642573842,
+      "step": 820
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 5e-05,
+      "loss": 0.4797,
+      "step": 902
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 0.6261923313140869,
+      "eval_per": 0.14098275699613683,
+      "eval_runtime": 116.5342,
+      "eval_samples_per_second": 6.976,
+      "eval_steps_per_second": 3.493,
+      "eval_wer": 0.16121506229294014,
+      "step": 902
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 0.3807,
+      "step": 984
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 0.5765030384063721,
+      "eval_per": 0.11716762461132574,
+      "eval_runtime": 116.5005,
+      "eval_samples_per_second": 6.979,
+      "eval_steps_per_second": 3.494,
+      "eval_wer": 0.13844430964490692,
+      "step": 984
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 0.3195,
+      "step": 1066
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 0.5666270852088928,
+      "eval_per": 0.09917082822952983,
+      "eval_runtime": 116.7786,
+      "eval_samples_per_second": 6.962,
+      "eval_steps_per_second": 3.485,
+      "eval_wer": 0.11912649899678036,
+      "step": 1066
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.2526,
+      "step": 1148
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 0.5758910775184631,
+      "eval_per": 0.09698011872232168,
+      "eval_runtime": 116.6256,
+      "eval_samples_per_second": 6.971,
+      "eval_steps_per_second": 3.49,
+      "eval_wer": 0.1165134618076618,
+      "step": 1148
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.2417,
+      "step": 1230
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 0.5459895730018616,
+      "eval_per": 0.09462451710166776,
+      "eval_runtime": 116.6754,
+      "eval_samples_per_second": 6.968,
+      "eval_steps_per_second": 3.488,
+      "eval_wer": 0.113807101861789,
+      "step": 1230
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.2072,
+      "step": 1312
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 0.5550991892814636,
+      "eval_per": 0.09116178271930651,
+      "eval_runtime": 116.9202,
+      "eval_samples_per_second": 6.953,
+      "eval_steps_per_second": 3.481,
+      "eval_wer": 0.10946759367271709,
+      "step": 1312
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.1881,
+      "step": 1394
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 0.5744515657424927,
+      "eval_per": 0.09172712710826345,
+      "eval_runtime": 116.943,
+      "eval_samples_per_second": 6.952,
+      "eval_steps_per_second": 3.48,
+      "eval_wer": 0.11016751434837385,
+      "step": 1394
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.1888,
+      "step": 1476
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 0.5731486678123474,
+      "eval_per": 0.09069066239517573,
+      "eval_runtime": 118.5572,
+      "eval_samples_per_second": 6.857,
+      "eval_steps_per_second": 3.433,
+      "eval_wer": 0.10942093229433998,
+      "step": 1476
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 5.555555555555556e-06,
+      "loss": 0.202,
+      "step": 1558
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 0.5773624777793884,
+      "eval_per": 0.08927730142278338,
+      "eval_runtime": 116.8858,
+      "eval_samples_per_second": 6.956,
+      "eval_steps_per_second": 3.482,
+      "eval_wer": 0.10806775232140357,
+      "step": 1558
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0,
+      "loss": 0.1813,
+      "step": 1640
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 0.5759608745574951,
+      "eval_per": 0.08915952134175069,
+      "eval_runtime": 117.5758,
+      "eval_samples_per_second": 6.915,
+      "eval_steps_per_second": 3.462,
+      "eval_wer": 0.10848770472679763,
+      "step": 1640
+    },
+    {
+      "epoch": 20.0,
+      "step": 1640,
+      "total_flos": 5.333828756305947e+17,
+      "train_loss": 2.0454907219584397,
+      "train_runtime": 3434.6913,
+      "train_samples_per_second": 0.949,
+      "train_steps_per_second": 0.477
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 1640,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "total_flos": 5.333828756305947e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebbd01c12a3d5ca4e1c0e7ab12d7d3b1506e8e466fa26cdb4eae4c79cab62ae2
+size 4600