--- 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-2 results: [] --- # hubert-classifier-aug-fold-2 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: 0.5616 - Accuracy: 0.8814 - Precision: 0.8965 - Recall: 0.8814 - F1: 0.8795 - Binary: 0.9182 ## 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 | 4.4205 | 0.0310 | 0.0078 | 0.0310 | 0.0104 | 0.2054 | | No log | 0.27 | 100 | 4.3291 | 0.0445 | 0.0108 | 0.0445 | 0.0126 | 0.2987 | | No log | 0.4 | 150 | 3.9426 | 0.0688 | 0.0295 | 0.0688 | 0.0221 | 0.3405 | | No log | 0.54 | 200 | 3.5950 | 0.0823 | 0.0420 | 0.0823 | 0.0336 | 0.3548 | | No log | 0.67 | 250 | 3.3440 | 0.1323 | 0.0736 | 0.1323 | 0.0675 | 0.3893 | | No log | 0.81 | 300 | 3.0808 | 0.2105 | 0.1599 | 0.2105 | 0.1353 | 0.4457 | | No log | 0.94 | 350 | 2.8041 | 0.3212 | 0.2060 | 0.3212 | 0.2180 | 0.5235 | | 3.8037 | 1.08 | 400 | 2.4494 | 0.3603 | 0.2731 | 0.3603 | 0.2746 | 0.5495 | | 3.8037 | 1.21 | 450 | 2.1029 | 0.4588 | 0.3806 | 0.4588 | 0.3860 | 0.6201 | | 3.8037 | 1.35 | 500 | 1.8222 | 0.5317 | 0.5092 | 0.5317 | 0.4811 | 0.6730 | | 3.8037 | 1.48 | 550 | 1.6391 | 0.5533 | 0.5401 | 0.5533 | 0.5040 | 0.6862 | | 3.8037 | 1.62 | 600 | 1.3955 | 0.6464 | 0.6432 | 0.6464 | 0.6129 | 0.7538 | | 3.8037 | 1.75 | 650 | 1.3006 | 0.6653 | 0.6734 | 0.6653 | 0.6399 | 0.7645 | | 3.8037 | 1.89 | 700 | 1.2028 | 0.6721 | 0.6946 | 0.6721 | 0.6492 | 0.7707 | | 1.9419 | 2.02 | 750 | 1.0903 | 0.6910 | 0.6900 | 0.6910 | 0.6661 | 0.7846 | | 1.9419 | 2.16 | 800 | 1.1952 | 0.7099 | 0.7198 | 0.7099 | 0.6923 | 0.7962 | | 1.9419 | 2.29 | 850 | 0.9361 | 0.7341 | 0.7547 | 0.7341 | 0.7222 | 0.8146 | | 1.9419 | 2.43 | 900 | 0.8789 | 0.7490 | 0.7725 | 0.7490 | 0.7348 | 0.8252 | | 1.9419 | 2.56 | 950 | 0.8519 | 0.7787 | 0.8015 | 0.7787 | 0.7714 | 0.8453 | | 1.9419 | 2.7 | 1000 | 0.8211 | 0.7692 | 0.7978 | 0.7692 | 0.7610 | 0.8394 | | 1.9419 | 2.83 | 1050 | 0.7343 | 0.7922 | 0.8128 | 0.7922 | 0.7849 | 0.8563 | | 1.9419 | 2.96 | 1100 | 0.8199 | 0.7760 | 0.8004 | 0.7760 | 0.7703 | 0.8451 | | 1.0676 | 3.1 | 1150 | 0.6783 | 0.8016 | 0.8161 | 0.8016 | 0.7956 | 0.8610 | | 1.0676 | 3.23 | 1200 | 0.7483 | 0.8043 | 0.8285 | 0.8043 | 0.7991 | 0.8638 | | 1.0676 | 3.37 | 1250 | 0.8469 | 0.7881 | 0.8090 | 0.7881 | 0.7795 | 0.8514 | | 1.0676 | 3.5 | 1300 | 0.7466 | 0.8003 | 0.8228 | 0.8003 | 0.7963 | 0.8603 | | 1.0676 | 3.64 | 1350 | 0.7441 | 0.8084 | 0.8375 | 0.8084 | 0.8059 | 0.8656 | | 1.0676 | 3.77 | 1400 | 0.6885 | 0.8124 | 0.8380 | 0.8124 | 0.8082 | 0.8695 | | 1.0676 | 3.91 | 1450 | 0.8319 | 0.7787 | 0.8045 | 0.7787 | 0.7729 | 0.8459 | | 0.75 | 4.04 | 1500 | 0.8044 | 0.8057 | 0.8320 | 0.8057 | 0.8017 | 0.8648 | | 0.75 | 4.18 | 1550 | 0.8120 | 0.8016 | 0.8230 | 0.8016 | 0.7964 | 0.8618 | | 0.75 | 4.31 | 1600 | 0.7503 | 0.8016 | 0.8166 | 0.8016 | 0.7965 | 0.8629 | | 0.75 | 4.45 | 1650 | 0.7646 | 0.8097 | 0.8269 | 0.8097 | 0.8025 | 0.8682 | | 0.75 | 4.58 | 1700 | 0.7328 | 0.8246 | 0.8442 | 0.8246 | 0.8189 | 0.8784 | | 0.75 | 4.72 | 1750 | 0.7019 | 0.8300 | 0.8479 | 0.8300 | 0.8270 | 0.8831 | | 0.75 | 4.85 | 1800 | 0.6364 | 0.8408 | 0.8558 | 0.8408 | 0.8379 | 0.8885 | | 0.75 | 4.99 | 1850 | 0.6562 | 0.8259 | 0.8461 | 0.8259 | 0.8214 | 0.8787 | | 0.5856 | 5.12 | 1900 | 0.6412 | 0.8340 | 0.8500 | 0.8340 | 0.8304 | 0.8843 | | 0.5856 | 5.26 | 1950 | 0.6739 | 0.8340 | 0.8548 | 0.8340 | 0.8333 | 0.8843 | | 0.5856 | 5.39 | 2000 | 0.7564 | 0.8097 | 0.8367 | 0.8097 | 0.8090 | 0.8673 | | 0.5856 | 5.53 | 2050 | 0.6495 | 0.8286 | 0.8473 | 0.8286 | 0.8264 | 0.8807 | | 0.5856 | 5.66 | 2100 | 0.7090 | 0.8165 | 0.8328 | 0.8165 | 0.8159 | 0.8717 | | 0.5856 | 5.8 | 2150 | 0.7712 | 0.8435 | 0.8654 | 0.8435 | 0.8419 | 0.8897 | | 0.5856 | 5.93 | 2200 | 0.7484 | 0.8273 | 0.8467 | 0.8273 | 0.8230 | 0.8802 | | 0.4896 | 6.06 | 2250 | 0.7999 | 0.8178 | 0.8344 | 0.8178 | 0.8163 | 0.8725 | | 0.4896 | 6.2 | 2300 | 0.7098 | 0.8300 | 0.8501 | 0.8300 | 0.8280 | 0.8821 | | 0.4896 | 6.33 | 2350 | 0.8158 | 0.8259 | 0.8434 | 0.8259 | 0.8226 | 0.8783 | | 0.4896 | 6.47 | 2400 | 0.6873 | 0.8381 | 0.8592 | 0.8381 | 0.8367 | 0.8868 | | 0.4896 | 6.6 | 2450 | 0.7824 | 0.8178 | 0.8394 | 0.8178 | 0.8156 | 0.8730 | | 0.4896 | 6.74 | 2500 | 0.6733 | 0.8448 | 0.8631 | 0.8448 | 0.8444 | 0.8915 | | 0.4896 | 6.87 | 2550 | 0.7658 | 0.8381 | 0.8577 | 0.8381 | 0.8347 | 0.8862 | | 0.4288 | 7.01 | 2600 | 0.7045 | 0.8367 | 0.8536 | 0.8367 | 0.8343 | 0.8853 | | 0.4288 | 7.14 | 2650 | 0.6885 | 0.8570 | 0.8748 | 0.8570 | 0.8545 | 0.8995 | | 0.4288 | 7.28 | 2700 | 0.7015 | 0.8475 | 0.8632 | 0.8475 | 0.8462 | 0.8934 | | 0.4288 | 7.41 | 2750 | 0.7385 | 0.8381 | 0.8539 | 0.8381 | 0.8363 | 0.8870 | | 0.4288 | 7.55 | 2800 | 0.7196 | 0.8475 | 0.8597 | 0.8475 | 0.8473 | 0.8930 | | 0.4288 | 7.68 | 2850 | 0.7285 | 0.8421 | 0.8558 | 0.8421 | 0.8399 | 0.8887 | | 0.4288 | 7.82 | 2900 | 0.7507 | 0.8367 | 0.8513 | 0.8367 | 0.8346 | 0.8849 | | 0.4288 | 7.95 | 2950 | 0.8049 | 0.8259 | 0.8522 | 0.8259 | 0.8225 | 0.8779 | | 0.3797 | 8.09 | 3000 | 0.7337 | 0.8529 | 0.8652 | 0.8529 | 0.8494 | 0.8978 | | 0.3797 | 8.22 | 3050 | 0.7414 | 0.8475 | 0.8628 | 0.8475 | 0.8451 | 0.8924 | | 0.3797 | 8.36 | 3100 | 0.8024 | 0.8394 | 0.8592 | 0.8394 | 0.8364 | 0.8877 | | 0.3797 | 8.49 | 3150 | 0.7642 | 0.8435 | 0.8635 | 0.8435 | 0.8405 | 0.8904 | | 0.3797 | 8.63 | 3200 | 0.7560 | 0.8448 | 0.8641 | 0.8448 | 0.8415 | 0.8906 | | 0.3797 | 8.76 | 3250 | 0.7889 | 0.8408 | 0.8640 | 0.8408 | 0.8373 | 0.8887 | | 0.3797 | 8.89 | 3300 | 0.7479 | 0.8448 | 0.8624 | 0.8448 | 0.8436 | 0.8915 | | 0.3486 | 9.03 | 3350 | 0.7430 | 0.8543 | 0.8727 | 0.8543 | 0.8529 | 0.8977 | | 0.3486 | 9.16 | 3400 | 0.7325 | 0.8394 | 0.8549 | 0.8394 | 0.8367 | 0.8877 | | 0.3486 | 9.3 | 3450 | 0.7623 | 0.8489 | 0.8641 | 0.8489 | 0.8441 | 0.8949 | | 0.3486 | 9.43 | 3500 | 0.7893 | 0.8462 | 0.8645 | 0.8462 | 0.8425 | 0.8924 | | 0.3486 | 9.57 | 3550 | 0.8721 | 0.8273 | 0.8504 | 0.8273 | 0.8239 | 0.8808 | | 0.3486 | 9.7 | 3600 | 0.7886 | 0.8489 | 0.8699 | 0.8489 | 0.8454 | 0.8939 | | 0.3486 | 9.84 | 3650 | 0.7844 | 0.8502 | 0.8692 | 0.8502 | 0.8469 | 0.8953 | | 0.3486 | 9.97 | 3700 | 0.9039 | 0.8232 | 0.8471 | 0.8232 | 0.8179 | 0.8769 | ### Framework versions - Transformers 4.38.2 - Pytorch 2.3.0 - Datasets 2.19.1 - Tokenizers 0.15.1