--- 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 results: [] --- # hubert-classifier-aug 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: 0.5366 - Accuracy: 0.8706 - Precision: 0.8951 - Recall: 0.8706 - F1: 0.8682 - Binary: 0.9078 ## 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 - num_epochs: 100 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Binary | |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:| | No log | 0.19 | 50 | 3.9567 | 0.0566 | 0.0103 | 0.0566 | 0.0141 | 0.3307 | | No log | 0.38 | 100 | 3.4847 | 0.0458 | 0.0044 | 0.0458 | 0.0076 | 0.3286 | | No log | 0.58 | 150 | 3.2446 | 0.0728 | 0.0102 | 0.0728 | 0.0173 | 0.3493 | | No log | 0.77 | 200 | 3.1428 | 0.1078 | 0.0310 | 0.1078 | 0.0409 | 0.3682 | | No log | 0.96 | 250 | 2.9478 | 0.1590 | 0.0774 | 0.1590 | 0.0809 | 0.4073 | | No log | 1.15 | 300 | 2.7731 | 0.2022 | 0.1264 | 0.2022 | 0.1206 | 0.4418 | | No log | 1.34 | 350 | 2.5200 | 0.2615 | 0.1732 | 0.2615 | 0.1606 | 0.4801 | | No log | 1.53 | 400 | 2.3855 | 0.3261 | 0.2286 | 0.3261 | 0.2292 | 0.5221 | | No log | 1.73 | 450 | 2.1667 | 0.3504 | 0.2931 | 0.3504 | 0.2719 | 0.5423 | | 3.2507 | 1.92 | 500 | 2.0399 | 0.4340 | 0.3626 | 0.4340 | 0.3608 | 0.6032 | | 3.2507 | 2.11 | 550 | 1.8119 | 0.4825 | 0.4560 | 0.4825 | 0.4230 | 0.6380 | | 3.2507 | 2.3 | 600 | 1.6704 | 0.5175 | 0.4294 | 0.5175 | 0.4487 | 0.6606 | | 3.2507 | 2.49 | 650 | 1.5691 | 0.5472 | 0.5141 | 0.5472 | 0.4955 | 0.6809 | | 3.2507 | 2.68 | 700 | 1.5136 | 0.6065 | 0.6009 | 0.6065 | 0.5648 | 0.7221 | | 3.2507 | 2.88 | 750 | 1.3633 | 0.6334 | 0.5725 | 0.6334 | 0.5806 | 0.7426 | | 3.2507 | 3.07 | 800 | 1.3163 | 0.6388 | 0.6464 | 0.6388 | 0.6012 | 0.7466 | | 3.2507 | 3.26 | 850 | 1.1000 | 0.7143 | 0.7039 | 0.7143 | 0.6741 | 0.7995 | | 3.2507 | 3.45 | 900 | 1.0805 | 0.7062 | 0.7037 | 0.7062 | 0.6691 | 0.7935 | | 3.2507 | 3.64 | 950 | 1.0359 | 0.7385 | 0.7487 | 0.7385 | 0.7102 | 0.8164 | | 2.027 | 3.84 | 1000 | 0.9199 | 0.7790 | 0.7804 | 0.7790 | 0.7508 | 0.8445 | | 2.027 | 4.03 | 1050 | 0.9748 | 0.7224 | 0.7335 | 0.7224 | 0.6918 | 0.8040 | | 2.027 | 4.22 | 1100 | 0.8482 | 0.7682 | 0.7788 | 0.7682 | 0.7518 | 0.8361 | | 2.027 | 4.41 | 1150 | 0.8507 | 0.7574 | 0.7642 | 0.7574 | 0.7358 | 0.8294 | | 2.027 | 4.6 | 1200 | 0.8277 | 0.7682 | 0.7852 | 0.7682 | 0.7507 | 0.8380 | | 2.027 | 4.79 | 1250 | 0.7315 | 0.7709 | 0.7923 | 0.7709 | 0.7542 | 0.8418 | | 2.027 | 4.99 | 1300 | 0.7434 | 0.7978 | 0.8366 | 0.7978 | 0.7873 | 0.8596 | | 2.027 | 5.18 | 1350 | 0.7260 | 0.8059 | 0.8242 | 0.8059 | 0.7930 | 0.8625 | | 2.027 | 5.37 | 1400 | 0.7265 | 0.7898 | 0.8087 | 0.7898 | 0.7725 | 0.8531 | | 2.027 | 5.56 | 1450 | 0.6691 | 0.8059 | 0.8280 | 0.8059 | 0.7968 | 0.8636 | | 1.4992 | 5.75 | 1500 | 0.6508 | 0.8167 | 0.8331 | 0.8167 | 0.8077 | 0.8709 | | 1.4992 | 5.94 | 1550 | 0.6404 | 0.8167 | 0.8325 | 0.8167 | 0.8084 | 0.8712 | | 1.4992 | 6.14 | 1600 | 0.6606 | 0.8140 | 0.8385 | 0.8140 | 0.8055 | 0.8682 | | 1.4992 | 6.33 | 1650 | 0.5687 | 0.8356 | 0.8416 | 0.8356 | 0.8222 | 0.8833 | | 1.4992 | 6.52 | 1700 | 0.5381 | 0.8410 | 0.8610 | 0.8410 | 0.8322 | 0.8889 | | 1.4992 | 6.71 | 1750 | 0.6056 | 0.8356 | 0.8628 | 0.8356 | 0.8289 | 0.8852 | | 1.4992 | 6.9 | 1800 | 0.5403 | 0.8491 | 0.8629 | 0.8491 | 0.8416 | 0.8954 | | 1.4992 | 7.09 | 1850 | 0.4901 | 0.8625 | 0.8773 | 0.8625 | 0.8576 | 0.9030 | | 1.4992 | 7.29 | 1900 | 0.5177 | 0.8544 | 0.8765 | 0.8544 | 0.8506 | 0.8965 | | 1.4992 | 7.48 | 1950 | 0.5830 | 0.8329 | 0.8658 | 0.8329 | 0.8245 | 0.8822 | | 1.2457 | 7.67 | 2000 | 0.4986 | 0.8491 | 0.8715 | 0.8491 | 0.8440 | 0.8946 | | 1.2457 | 7.86 | 2050 | 0.6022 | 0.8113 | 0.8488 | 0.8113 | 0.8089 | 0.8693 | | 1.2457 | 8.05 | 2100 | 0.5857 | 0.8383 | 0.8613 | 0.8383 | 0.8348 | 0.8871 | | 1.2457 | 8.25 | 2150 | 0.5669 | 0.8194 | 0.8485 | 0.8194 | 0.8194 | 0.8752 | | 1.2457 | 8.44 | 2200 | 0.5661 | 0.8571 | 0.8764 | 0.8571 | 0.8555 | 0.9003 | | 1.2457 | 8.63 | 2250 | 0.5170 | 0.8598 | 0.8919 | 0.8598 | 0.8547 | 0.9022 | | 1.2457 | 8.82 | 2300 | 0.5744 | 0.8248 | 0.8493 | 0.8248 | 0.8208 | 0.8757 | | 1.2457 | 9.01 | 2350 | 0.5577 | 0.8410 | 0.8650 | 0.8410 | 0.8344 | 0.8871 | | 1.2457 | 9.2 | 2400 | 0.5493 | 0.8275 | 0.8429 | 0.8275 | 0.8228 | 0.8784 | | 1.2457 | 9.4 | 2450 | 0.4822 | 0.8679 | 0.8913 | 0.8679 | 0.8654 | 0.9078 | | 1.0978 | 9.59 | 2500 | 0.4880 | 0.8464 | 0.8627 | 0.8464 | 0.8405 | 0.8938 | | 1.0978 | 9.78 | 2550 | 0.5233 | 0.8625 | 0.8771 | 0.8625 | 0.8520 | 0.9038 | | 1.0978 | 9.97 | 2600 | 0.4864 | 0.8733 | 0.8903 | 0.8733 | 0.8693 | 0.9108 | | 1.0978 | 10.16 | 2650 | 0.5167 | 0.8706 | 0.8932 | 0.8706 | 0.8649 | 0.9086 | | 1.0978 | 10.35 | 2700 | 0.4831 | 0.8706 | 0.8872 | 0.8706 | 0.8676 | 0.9086 | | 1.0978 | 10.55 | 2750 | 0.4824 | 0.8760 | 0.8982 | 0.8760 | 0.8741 | 0.9132 | | 1.0978 | 10.74 | 2800 | 0.5156 | 0.8598 | 0.8850 | 0.8598 | 0.8561 | 0.9011 | | 1.0978 | 10.93 | 2850 | 0.5065 | 0.8895 | 0.9124 | 0.8895 | 0.8873 | 0.9210 | | 1.0978 | 11.12 | 2900 | 0.4637 | 0.8787 | 0.8990 | 0.8787 | 0.8772 | 0.9143 | | 1.0978 | 11.31 | 2950 | 0.4574 | 0.8922 | 0.9056 | 0.8922 | 0.8908 | 0.9232 | | 0.9986 | 11.51 | 3000 | 0.5472 | 0.8760 | 0.9029 | 0.8760 | 0.8755 | 0.9124 | | 0.9986 | 11.7 | 3050 | 0.5353 | 0.8679 | 0.8911 | 0.8679 | 0.8642 | 0.9108 | | 0.9986 | 11.89 | 3100 | 0.4301 | 0.8679 | 0.8818 | 0.8679 | 0.8617 | 0.9067 | | 0.9986 | 12.08 | 3150 | 0.5122 | 0.8544 | 0.8746 | 0.8544 | 0.8520 | 0.8957 | | 0.9986 | 12.27 | 3200 | 0.4837 | 0.8922 | 0.9080 | 0.8922 | 0.8888 | 0.9229 | | 0.9986 | 12.46 | 3250 | 0.5032 | 0.8706 | 0.8908 | 0.8706 | 0.8669 | 0.9078 | | 0.9986 | 12.66 | 3300 | 0.5752 | 0.8544 | 0.8710 | 0.8544 | 0.8479 | 0.8957 | | 0.9986 | 12.85 | 3350 | 0.6008 | 0.8491 | 0.8737 | 0.8491 | 0.8428 | 0.8935 | | 0.9986 | 13.04 | 3400 | 0.4820 | 0.8733 | 0.8960 | 0.8733 | 0.8701 | 0.9127 | | 0.9986 | 13.23 | 3450 | 0.5366 | 0.8706 | 0.8951 | 0.8706 | 0.8682 | 0.9078 | ### Framework versions - Transformers 4.38.2 - Pytorch 2.3.0 - Datasets 2.19.1 - Tokenizers 0.15.1