--- 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-1 results: [] --- # hubert-classifier-aug-fold-1 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: 0.6112 - Accuracy: 0.8733 - Precision: 0.8849 - Recall: 0.8733 - F1: 0.8715 - Binary: 0.9115 ## 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 | 0.24 | 50 | 4.4173 | 0.0120 | 0.0018 | 0.0120 | 0.0027 | 0.1454 | | No log | 0.48 | 100 | 4.3112 | 0.0307 | 0.0029 | 0.0307 | 0.0050 | 0.2561 | | No log | 0.72 | 150 | 3.9716 | 0.0577 | 0.0136 | 0.0577 | 0.0137 | 0.3354 | | No log | 0.96 | 200 | 3.6532 | 0.0906 | 0.0647 | 0.0906 | 0.0408 | 0.3616 | | 4.2325 | 1.2 | 250 | 3.3860 | 0.1311 | 0.0767 | 0.1311 | 0.0725 | 0.3903 | | 4.2325 | 1.44 | 300 | 3.1896 | 0.2150 | 0.1277 | 0.2150 | 0.1379 | 0.4468 | | 4.2325 | 1.68 | 350 | 2.9240 | 0.2412 | 0.1475 | 0.2412 | 0.1486 | 0.4669 | | 4.2325 | 1.92 | 400 | 2.6191 | 0.2861 | 0.2519 | 0.2861 | 0.2143 | 0.4985 | | 3.2742 | 2.16 | 450 | 2.3504 | 0.3603 | 0.2949 | 0.3603 | 0.2791 | 0.5510 | | 3.2742 | 2.4 | 500 | 2.0177 | 0.4981 | 0.4172 | 0.4981 | 0.4130 | 0.6467 | | 3.2742 | 2.63 | 550 | 1.9152 | 0.5146 | 0.5098 | 0.5146 | 0.4630 | 0.6586 | | 3.2742 | 2.87 | 600 | 1.6539 | 0.5918 | 0.5981 | 0.5918 | 0.5415 | 0.7128 | | 2.3027 | 3.11 | 650 | 1.4801 | 0.6494 | 0.6389 | 0.6494 | 0.6128 | 0.7532 | | 2.3027 | 3.35 | 700 | 1.2164 | 0.7124 | 0.6887 | 0.7124 | 0.6790 | 0.7980 | | 2.3027 | 3.59 | 750 | 1.1214 | 0.7236 | 0.7205 | 0.7236 | 0.6985 | 0.8057 | | 2.3027 | 3.83 | 800 | 1.0199 | 0.7438 | 0.7357 | 0.7438 | 0.7187 | 0.8209 | | 1.6257 | 4.07 | 850 | 0.9595 | 0.7528 | 0.7644 | 0.7528 | 0.7354 | 0.8270 | | 1.6257 | 4.31 | 900 | 0.8867 | 0.7670 | 0.7720 | 0.7670 | 0.7507 | 0.8369 | | 1.6257 | 4.55 | 950 | 0.8603 | 0.7820 | 0.7875 | 0.7820 | 0.7713 | 0.8480 | | 1.6257 | 4.79 | 1000 | 0.7999 | 0.7723 | 0.7874 | 0.7723 | 0.7638 | 0.8413 | | 1.2686 | 5.03 | 1050 | 0.7813 | 0.7948 | 0.8123 | 0.7948 | 0.7873 | 0.8577 | | 1.2686 | 5.27 | 1100 | 0.7312 | 0.8165 | 0.8300 | 0.8165 | 0.8100 | 0.8709 | | 1.2686 | 5.51 | 1150 | 0.7178 | 0.8180 | 0.8347 | 0.8180 | 0.8132 | 0.8718 | | 1.2686 | 5.75 | 1200 | 0.7108 | 0.8060 | 0.8199 | 0.8060 | 0.8001 | 0.8646 | | 1.2686 | 5.99 | 1250 | 0.6504 | 0.8247 | 0.8304 | 0.8247 | 0.8165 | 0.8772 | | 1.0234 | 6.23 | 1300 | 0.6944 | 0.8187 | 0.8310 | 0.8187 | 0.8125 | 0.8725 | | 1.0234 | 6.47 | 1350 | 0.6046 | 0.8397 | 0.8548 | 0.8397 | 0.8383 | 0.8880 | | 1.0234 | 6.71 | 1400 | 0.6195 | 0.8382 | 0.8489 | 0.8382 | 0.8335 | 0.8869 | | 1.0234 | 6.95 | 1450 | 0.6295 | 0.8412 | 0.8514 | 0.8412 | 0.8372 | 0.8894 | | 0.8831 | 7.19 | 1500 | 0.6205 | 0.8337 | 0.8419 | 0.8337 | 0.8303 | 0.8837 | | 0.8831 | 7.43 | 1550 | 0.6006 | 0.8464 | 0.8590 | 0.8464 | 0.8447 | 0.8935 | | 0.8831 | 7.66 | 1600 | 0.5860 | 0.8592 | 0.8684 | 0.8592 | 0.8579 | 0.9036 | | 0.8831 | 7.9 | 1650 | 0.5906 | 0.8419 | 0.8525 | 0.8419 | 0.8409 | 0.8909 | | 0.7822 | 8.14 | 1700 | 0.6277 | 0.8457 | 0.8567 | 0.8457 | 0.8420 | 0.8922 | | 0.7822 | 8.38 | 1750 | 0.5977 | 0.8532 | 0.8659 | 0.8532 | 0.8496 | 0.8980 | | 0.7822 | 8.62 | 1800 | 0.5970 | 0.8622 | 0.8696 | 0.8622 | 0.8601 | 0.9037 | | 0.7822 | 8.86 | 1850 | 0.5471 | 0.8607 | 0.8678 | 0.8607 | 0.8593 | 0.9034 | | 0.7039 | 9.1 | 1900 | 0.5848 | 0.8569 | 0.8687 | 0.8569 | 0.8541 | 0.8999 | | 0.7039 | 9.34 | 1950 | 0.5518 | 0.8682 | 0.8748 | 0.8682 | 0.8665 | 0.9082 | | 0.7039 | 9.58 | 2000 | 0.5860 | 0.8667 | 0.8760 | 0.8667 | 0.8653 | 0.9069 | | 0.7039 | 9.82 | 2050 | 0.5937 | 0.8652 | 0.8743 | 0.8652 | 0.8624 | 0.9053 | | 0.6314 | 10.06 | 2100 | 0.5993 | 0.8607 | 0.8688 | 0.8607 | 0.8592 | 0.9021 | | 0.6314 | 10.3 | 2150 | 0.5401 | 0.8697 | 0.8780 | 0.8697 | 0.8675 | 0.9094 | | 0.6314 | 10.54 | 2200 | 0.5701 | 0.8607 | 0.8744 | 0.8607 | 0.8600 | 0.9026 | | 0.6314 | 10.78 | 2250 | 0.5303 | 0.8757 | 0.8854 | 0.8757 | 0.8738 | 0.9129 | | 0.6017 | 11.02 | 2300 | 0.5408 | 0.8772 | 0.8830 | 0.8772 | 0.8752 | 0.9139 | | 0.6017 | 11.26 | 2350 | 0.5218 | 0.8809 | 0.8857 | 0.8809 | 0.8785 | 0.9168 | | 0.6017 | 11.5 | 2400 | 0.6290 | 0.8584 | 0.8694 | 0.8584 | 0.8555 | 0.9005 | | 0.6017 | 11.74 | 2450 | 0.5580 | 0.8644 | 0.8715 | 0.8644 | 0.8631 | 0.9055 | | 0.6017 | 11.98 | 2500 | 0.5415 | 0.8652 | 0.8722 | 0.8652 | 0.8641 | 0.9060 | | 0.5539 | 12.22 | 2550 | 0.5297 | 0.8749 | 0.8835 | 0.8749 | 0.8738 | 0.9123 | | 0.5539 | 12.46 | 2600 | 0.5721 | 0.8682 | 0.8765 | 0.8682 | 0.8659 | 0.9079 | | 0.5539 | 12.69 | 2650 | 0.5989 | 0.8697 | 0.8802 | 0.8697 | 0.8689 | 0.9098 | | 0.5539 | 12.93 | 2700 | 0.6499 | 0.8629 | 0.8757 | 0.8629 | 0.8613 | 0.9053 | | 0.5168 | 13.17 | 2750 | 0.5816 | 0.8749 | 0.8831 | 0.8749 | 0.8739 | 0.9124 | | 0.5168 | 13.41 | 2800 | 0.6052 | 0.8764 | 0.8868 | 0.8764 | 0.8746 | 0.9133 | | 0.5168 | 13.65 | 2850 | 0.6148 | 0.8697 | 0.8803 | 0.8697 | 0.8679 | 0.9084 | | 0.5168 | 13.89 | 2900 | 0.6010 | 0.8779 | 0.8875 | 0.8779 | 0.8764 | 0.9153 | | 0.4881 | 14.13 | 2950 | 0.5583 | 0.8801 | 0.8893 | 0.8801 | 0.8790 | 0.9160 | | 0.4881 | 14.37 | 3000 | 0.5880 | 0.8779 | 0.8859 | 0.8779 | 0.8763 | 0.9154 | | 0.4881 | 14.61 | 3050 | 0.5560 | 0.8794 | 0.8893 | 0.8794 | 0.8774 | 0.9169 | | 0.4881 | 14.85 | 3100 | 0.5339 | 0.8831 | 0.8896 | 0.8831 | 0.8813 | 0.9191 | | 0.4611 | 15.09 | 3150 | 0.5541 | 0.8816 | 0.8869 | 0.8816 | 0.8803 | 0.9176 | | 0.4611 | 15.33 | 3200 | 0.5848 | 0.8839 | 0.8900 | 0.8839 | 0.8822 | 0.9190 | | 0.4611 | 15.57 | 3250 | 0.5712 | 0.8869 | 0.8924 | 0.8869 | 0.8862 | 0.9207 | | 0.4611 | 15.81 | 3300 | 0.5159 | 0.8921 | 0.8983 | 0.8921 | 0.8916 | 0.9246 | | 0.4345 | 16.05 | 3350 | 0.5486 | 0.8839 | 0.8920 | 0.8839 | 0.8834 | 0.9191 | | 0.4345 | 16.29 | 3400 | 0.5568 | 0.8816 | 0.8882 | 0.8816 | 0.8806 | 0.9179 | | 0.4345 | 16.53 | 3450 | 0.5752 | 0.8839 | 0.8896 | 0.8839 | 0.8828 | 0.9186 | | 0.4345 | 16.77 | 3500 | 0.5716 | 0.8831 | 0.8897 | 0.8831 | 0.8814 | 0.9181 | | 0.4208 | 17.01 | 3550 | 0.5562 | 0.8816 | 0.8906 | 0.8816 | 0.8808 | 0.9170 | | 0.4208 | 17.25 | 3600 | 0.5623 | 0.8809 | 0.8881 | 0.8809 | 0.8804 | 0.9165 | | 0.4208 | 17.49 | 3650 | 0.5756 | 0.8914 | 0.8982 | 0.8914 | 0.8910 | 0.9238 | | 0.4208 | 17.72 | 3700 | 0.5662 | 0.8861 | 0.8915 | 0.8861 | 0.8849 | 0.9199 | | 0.4208 | 17.96 | 3750 | 0.5965 | 0.8891 | 0.8952 | 0.8891 | 0.8882 | 0.9220 | | 0.4137 | 18.2 | 3800 | 0.5827 | 0.8876 | 0.8958 | 0.8876 | 0.8871 | 0.9217 | | 0.4137 | 18.44 | 3850 | 0.5463 | 0.8929 | 0.8998 | 0.8929 | 0.8923 | 0.9249 | | 0.4137 | 18.68 | 3900 | 0.5731 | 0.8869 | 0.8932 | 0.8869 | 0.8858 | 0.9207 | | 0.4137 | 18.92 | 3950 | 0.5538 | 0.8869 | 0.8933 | 0.8869 | 0.8853 | 0.9209 | | 0.39 | 19.16 | 4000 | 0.5692 | 0.8869 | 0.8934 | 0.8869 | 0.8854 | 0.9209 | | 0.39 | 19.4 | 4050 | 0.5288 | 0.8944 | 0.8998 | 0.8944 | 0.8934 | 0.9259 | | 0.39 | 19.64 | 4100 | 0.5907 | 0.8884 | 0.8951 | 0.8884 | 0.8879 | 0.9219 | | 0.39 | 19.88 | 4150 | 0.5595 | 0.8884 | 0.8963 | 0.8884 | 0.8869 | 0.9219 | | 0.359 | 20.12 | 4200 | 0.6029 | 0.8779 | 0.8868 | 0.8779 | 0.8776 | 0.9141 | | 0.359 | 20.36 | 4250 | 0.5650 | 0.8959 | 0.9026 | 0.8959 | 0.8956 | 0.9272 | | 0.359 | 20.6 | 4300 | 0.5699 | 0.8869 | 0.8935 | 0.8869 | 0.8863 | 0.9211 | | 0.359 | 20.84 | 4350 | 0.5717 | 0.8816 | 0.8884 | 0.8816 | 0.8809 | 0.9172 | | 0.3685 | 21.08 | 4400 | 0.5991 | 0.8794 | 0.8878 | 0.8794 | 0.8780 | 0.9157 | | 0.3685 | 21.32 | 4450 | 0.5760 | 0.8959 | 0.9037 | 0.8959 | 0.8954 | 0.9274 | | 0.3685 | 21.56 | 4500 | 0.5753 | 0.8974 | 0.9047 | 0.8974 | 0.8966 | 0.9285 | | 0.3685 | 21.8 | 4550 | 0.5693 | 0.8891 | 0.8959 | 0.8891 | 0.8873 | 0.9227 | | 0.3472 | 22.04 | 4600 | 0.5866 | 0.8831 | 0.8905 | 0.8831 | 0.8820 | 0.9184 | | 0.3472 | 22.28 | 4650 | 0.5781 | 0.8899 | 0.8969 | 0.8899 | 0.8892 | 0.9233 | | 0.3472 | 22.51 | 4700 | 0.6050 | 0.8921 | 0.8989 | 0.8921 | 0.8910 | 0.9240 | | 0.3472 | 22.75 | 4750 | 0.5826 | 0.8914 | 0.8965 | 0.8914 | 0.8906 | 0.9238 | | 0.3472 | 22.99 | 4800 | 0.5809 | 0.8981 | 0.9050 | 0.8981 | 0.8973 | 0.9285 | | 0.3316 | 23.23 | 4850 | 0.6249 | 0.8869 | 0.8942 | 0.8869 | 0.8865 | 0.9210 | | 0.3316 | 23.47 | 4900 | 0.5971 | 0.8876 | 0.8937 | 0.8876 | 0.8869 | 0.9214 | | 0.3316 | 23.71 | 4950 | 0.5849 | 0.8884 | 0.8948 | 0.8884 | 0.8883 | 0.9217 | | 0.3316 | 23.95 | 5000 | 0.5806 | 0.8854 | 0.8913 | 0.8854 | 0.8854 | 0.9199 | | 0.3066 | 24.19 | 5050 | 0.5833 | 0.8936 | 0.8996 | 0.8936 | 0.8929 | 0.9254 | | 0.3066 | 24.43 | 5100 | 0.5802 | 0.8966 | 0.9033 | 0.8966 | 0.8963 | 0.9275 | | 0.3066 | 24.67 | 5150 | 0.5742 | 0.8906 | 0.8971 | 0.8906 | 0.8901 | 0.9233 | ### Framework versions - Transformers 4.38.2 - Pytorch 2.3.0 - Datasets 2.19.1 - Tokenizers 0.15.1