--- license: apache-2.0 base_model: facebook/hubert-base-ls960 tags: - generated_from_trainer metrics: - accuracy - precision - recall - f1 model-index: - name: hubert-classifier-aug-fold-0 results: [] --- # hubert-classifier-aug-fold-0 This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.7030 - Accuracy: 0.8571 - Precision: 0.8685 - Recall: 0.8571 - F1: 0.8514 - Binary: 0.9007 ## 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: 32 - eval_batch_size: 32 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 128 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Binary | |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:| | No log | 0.13 | 50 | 4.4206 | 0.0243 | 0.0145 | 0.0243 | 0.0115 | 0.1888 | | No log | 0.27 | 100 | 4.3046 | 0.0458 | 0.0194 | 0.0458 | 0.0171 | 0.3013 | | No log | 0.4 | 150 | 3.9493 | 0.0714 | 0.0250 | 0.0714 | 0.0284 | 0.3407 | | No log | 0.54 | 200 | 3.7625 | 0.0768 | 0.0518 | 0.0768 | 0.0328 | 0.3255 | | No log | 0.67 | 250 | 3.3324 | 0.1644 | 0.0956 | 0.1644 | 0.0982 | 0.4112 | | No log | 0.81 | 300 | 3.0658 | 0.2466 | 0.1774 | 0.2466 | 0.1719 | 0.4704 | | No log | 0.94 | 350 | 2.7642 | 0.3450 | 0.2437 | 0.3450 | 0.2520 | 0.5375 | | 3.7949 | 1.08 | 400 | 2.4739 | 0.3760 | 0.3261 | 0.3760 | 0.3038 | 0.5586 | | 3.7949 | 1.21 | 450 | 2.1757 | 0.4757 | 0.3860 | 0.4757 | 0.4022 | 0.6303 | | 3.7949 | 1.35 | 500 | 1.8422 | 0.5323 | 0.4744 | 0.5323 | 0.4677 | 0.6717 | | 3.7949 | 1.48 | 550 | 1.6818 | 0.5889 | 0.5464 | 0.5889 | 0.5402 | 0.7100 | | 3.7949 | 1.62 | 600 | 1.4944 | 0.6173 | 0.6054 | 0.6173 | 0.5761 | 0.7319 | | 3.7949 | 1.75 | 650 | 1.3503 | 0.6429 | 0.6457 | 0.6429 | 0.6069 | 0.7481 | | 3.7949 | 1.89 | 700 | 1.2159 | 0.6752 | 0.6724 | 0.6752 | 0.6372 | 0.7732 | | 1.9673 | 2.02 | 750 | 1.0849 | 0.7049 | 0.7117 | 0.7049 | 0.6748 | 0.7930 | | 1.9673 | 2.16 | 800 | 1.0424 | 0.7183 | 0.7257 | 0.7183 | 0.6987 | 0.8011 | | 1.9673 | 2.29 | 850 | 0.8607 | 0.7830 | 0.7911 | 0.7830 | 0.7694 | 0.8470 | | 1.9673 | 2.43 | 900 | 0.8606 | 0.7668 | 0.7813 | 0.7668 | 0.7506 | 0.8372 | | 1.9673 | 2.56 | 950 | 0.7939 | 0.7803 | 0.7739 | 0.7803 | 0.7606 | 0.8487 | | 1.9673 | 2.7 | 1000 | 0.7883 | 0.8059 | 0.8257 | 0.8059 | 0.7976 | 0.8656 | | 1.9673 | 2.83 | 1050 | 0.7567 | 0.8032 | 0.8214 | 0.8032 | 0.7947 | 0.8640 | | 1.9673 | 2.97 | 1100 | 0.6989 | 0.8181 | 0.8399 | 0.8181 | 0.8063 | 0.8745 | | 1.0987 | 3.1 | 1150 | 0.7500 | 0.8100 | 0.8223 | 0.8100 | 0.8043 | 0.8660 | | 1.0987 | 3.24 | 1200 | 0.6802 | 0.8261 | 0.8381 | 0.8261 | 0.8184 | 0.8794 | | 1.0987 | 3.37 | 1250 | 0.6614 | 0.8396 | 0.8558 | 0.8396 | 0.8359 | 0.8877 | | 1.0987 | 3.51 | 1300 | 0.6928 | 0.8261 | 0.8511 | 0.8261 | 0.8236 | 0.8791 | | 1.0987 | 3.64 | 1350 | 0.6146 | 0.8410 | 0.8588 | 0.8410 | 0.8401 | 0.8896 | | 1.0987 | 3.78 | 1400 | 0.6958 | 0.8248 | 0.8412 | 0.8248 | 0.8191 | 0.8796 | | 1.0987 | 3.91 | 1450 | 0.6785 | 0.8342 | 0.8556 | 0.8342 | 0.8309 | 0.8857 | | 0.7483 | 4.05 | 1500 | 0.7412 | 0.8261 | 0.8461 | 0.8261 | 0.8244 | 0.8784 | | 0.7483 | 4.18 | 1550 | 0.6778 | 0.8356 | 0.8538 | 0.8356 | 0.8317 | 0.8868 | | 0.7483 | 4.32 | 1600 | 0.7032 | 0.8437 | 0.8657 | 0.8437 | 0.8405 | 0.8946 | | 0.7483 | 4.45 | 1650 | 0.7373 | 0.8329 | 0.8564 | 0.8329 | 0.8299 | 0.8850 | | 0.7483 | 4.59 | 1700 | 0.6958 | 0.8423 | 0.8593 | 0.8423 | 0.8401 | 0.8915 | | 0.7483 | 4.72 | 1750 | 0.7395 | 0.8329 | 0.8513 | 0.8329 | 0.8327 | 0.8865 | | 0.7483 | 4.86 | 1800 | 0.7017 | 0.8477 | 0.8651 | 0.8477 | 0.8453 | 0.8953 | | 0.7483 | 4.99 | 1850 | 0.7240 | 0.8423 | 0.8582 | 0.8423 | 0.8410 | 0.8922 | | 0.5887 | 5.12 | 1900 | 0.6810 | 0.8464 | 0.8694 | 0.8464 | 0.8451 | 0.8943 | | 0.5887 | 5.26 | 1950 | 0.6091 | 0.8706 | 0.8828 | 0.8706 | 0.8688 | 0.9111 | | 0.5887 | 5.39 | 2000 | 0.6617 | 0.8491 | 0.8669 | 0.8491 | 0.8474 | 0.8974 | | 0.5887 | 5.53 | 2050 | 0.6712 | 0.8477 | 0.8662 | 0.8477 | 0.8458 | 0.8966 | | 0.5887 | 5.66 | 2100 | 0.6988 | 0.8437 | 0.8570 | 0.8437 | 0.8413 | 0.8915 | | 0.5887 | 5.8 | 2150 | 0.6644 | 0.8477 | 0.8624 | 0.8477 | 0.8455 | 0.8953 | | 0.5887 | 5.93 | 2200 | 0.6416 | 0.8652 | 0.8790 | 0.8652 | 0.8622 | 0.9073 | | 0.485 | 6.07 | 2250 | 0.6484 | 0.8585 | 0.8705 | 0.8585 | 0.8568 | 0.9030 | | 0.485 | 6.2 | 2300 | 0.6690 | 0.8585 | 0.8736 | 0.8585 | 0.8564 | 0.9019 | | 0.485 | 6.34 | 2350 | 0.6469 | 0.8639 | 0.8790 | 0.8639 | 0.8616 | 0.9071 | | 0.485 | 6.47 | 2400 | 0.7418 | 0.8518 | 0.8684 | 0.8518 | 0.8515 | 0.8968 | | 0.485 | 6.61 | 2450 | 0.6821 | 0.8625 | 0.8788 | 0.8625 | 0.8615 | 0.9075 | | 0.485 | 6.74 | 2500 | 0.7012 | 0.8652 | 0.8837 | 0.8652 | 0.8636 | 0.9090 | | 0.485 | 6.88 | 2550 | 0.6546 | 0.8652 | 0.8819 | 0.8652 | 0.8658 | 0.9085 | | 0.4243 | 7.01 | 2600 | 0.6619 | 0.8639 | 0.8769 | 0.8639 | 0.8631 | 0.9070 | | 0.4243 | 7.15 | 2650 | 0.7000 | 0.8531 | 0.8716 | 0.8531 | 0.8518 | 0.8995 | | 0.4243 | 7.28 | 2700 | 0.6560 | 0.8720 | 0.8864 | 0.8720 | 0.8715 | 0.9112 | | 0.4243 | 7.42 | 2750 | 0.6458 | 0.8639 | 0.8786 | 0.8639 | 0.8634 | 0.9074 | | 0.4243 | 7.55 | 2800 | 0.6701 | 0.8747 | 0.8896 | 0.8747 | 0.8742 | 0.9155 | | 0.4243 | 7.69 | 2850 | 0.7282 | 0.8477 | 0.8706 | 0.8477 | 0.8477 | 0.8970 | | 0.4243 | 7.82 | 2900 | 0.6578 | 0.8612 | 0.8726 | 0.8612 | 0.8597 | 0.9061 | | 0.4243 | 7.96 | 2950 | 0.6244 | 0.8720 | 0.8829 | 0.8720 | 0.8704 | 0.9142 | | 0.378 | 8.09 | 3000 | 0.6445 | 0.8733 | 0.8896 | 0.8733 | 0.8726 | 0.9140 | | 0.378 | 8.23 | 3050 | 0.6983 | 0.8612 | 0.8766 | 0.8612 | 0.8606 | 0.9055 | | 0.378 | 8.36 | 3100 | 0.6355 | 0.8760 | 0.8922 | 0.8760 | 0.8750 | 0.9154 | | 0.378 | 8.5 | 3150 | 0.6770 | 0.8747 | 0.8883 | 0.8747 | 0.8726 | 0.9135 | | 0.378 | 8.63 | 3200 | 0.6472 | 0.8706 | 0.8798 | 0.8706 | 0.8680 | 0.9097 | | 0.378 | 8.77 | 3250 | 0.7739 | 0.8544 | 0.8691 | 0.8544 | 0.8512 | 0.8970 | | 0.378 | 8.9 | 3300 | 0.6805 | 0.8612 | 0.8766 | 0.8612 | 0.8587 | 0.9046 | | 0.3491 | 9.04 | 3350 | 0.6382 | 0.8733 | 0.8829 | 0.8733 | 0.8717 | 0.9135 | | 0.3491 | 9.17 | 3400 | 0.6927 | 0.8652 | 0.8793 | 0.8652 | 0.8642 | 0.9080 | | 0.3491 | 9.31 | 3450 | 0.8407 | 0.8518 | 0.8711 | 0.8518 | 0.8488 | 0.8996 | | 0.3491 | 9.44 | 3500 | 0.6628 | 0.8733 | 0.8823 | 0.8733 | 0.8714 | 0.9121 | ### Framework versions - Transformers 4.38.2 - Pytorch 2.3.0 - Datasets 2.19.1 - Tokenizers 0.15.1