<!-- 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. -->
my_MFCC_VITmodelBIT1
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3329
- Accuracy: 0.8909
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.6858 | 0.95 | 10 | 0.6256 | 0.8 |
0.5939 | 2.0 | 21 | 0.5227 | 0.8 |
0.561 | 2.95 | 31 | 0.4967 | 0.8 |
0.5175 | 4.0 | 42 | 0.4687 | 0.8 |
0.5334 | 4.95 | 52 | 0.4554 | 0.8 |
0.5074 | 6.0 | 63 | 0.4505 | 0.8 |
0.4852 | 6.95 | 73 | 0.4422 | 0.8 |
0.4711 | 8.0 | 84 | 0.4033 | 0.8 |
0.4636 | 8.95 | 94 | 0.4193 | 0.8242 |
0.5 | 10.0 | 105 | 0.3682 | 0.8485 |
0.4255 | 10.95 | 115 | 0.4124 | 0.7879 |
0.4258 | 12.0 | 126 | 0.4144 | 0.8364 |
0.4542 | 12.95 | 136 | 0.3729 | 0.8303 |
0.3631 | 14.0 | 147 | 0.4177 | 0.8303 |
0.4919 | 14.95 | 157 | 0.3634 | 0.8303 |
0.405 | 16.0 | 168 | 0.3081 | 0.8970 |
0.3908 | 16.95 | 178 | 0.3965 | 0.8424 |
0.4064 | 18.0 | 189 | 0.3502 | 0.8364 |
0.345 | 18.95 | 199 | 0.3427 | 0.8303 |
0.363 | 20.0 | 210 | 0.2901 | 0.8909 |
0.3278 | 20.95 | 220 | 0.3289 | 0.8667 |
0.3074 | 22.0 | 231 | 0.3593 | 0.8121 |
0.3469 | 22.95 | 241 | 0.2968 | 0.8727 |
0.3545 | 24.0 | 252 | 0.4895 | 0.7394 |
0.3457 | 24.95 | 262 | 0.3278 | 0.8788 |
0.339 | 26.0 | 273 | 0.3363 | 0.8424 |
0.3023 | 26.95 | 283 | 0.3420 | 0.8667 |
0.3462 | 28.0 | 294 | 0.3377 | 0.8364 |
0.2999 | 28.95 | 304 | 0.3599 | 0.8606 |
0.2713 | 30.0 | 315 | 0.3054 | 0.8727 |
0.2805 | 30.95 | 325 | 0.3414 | 0.8424 |
0.294 | 32.0 | 336 | 0.2949 | 0.8788 |
0.2884 | 32.95 | 346 | 0.2989 | 0.8545 |
0.2936 | 34.0 | 357 | 0.3898 | 0.8424 |
0.3077 | 34.95 | 367 | 0.3450 | 0.8545 |
0.3316 | 36.0 | 378 | 0.2584 | 0.9152 |
0.2769 | 36.95 | 388 | 0.2774 | 0.8788 |
0.2555 | 38.0 | 399 | 0.3349 | 0.8303 |
0.2512 | 38.95 | 409 | 0.3747 | 0.8545 |
0.2707 | 40.0 | 420 | 0.3558 | 0.8303 |
0.2638 | 40.95 | 430 | 0.3931 | 0.7939 |
0.2746 | 42.0 | 441 | 0.3997 | 0.8242 |
0.307 | 42.95 | 451 | 0.3194 | 0.8485 |
0.2269 | 44.0 | 462 | 0.4378 | 0.8182 |
0.2142 | 44.95 | 472 | 0.3499 | 0.8424 |
0.2102 | 46.0 | 483 | 0.3766 | 0.8303 |
0.247 | 46.95 | 493 | 0.3521 | 0.8242 |
0.2347 | 48.0 | 504 | 0.3583 | 0.8667 |
0.2081 | 48.95 | 514 | 0.3162 | 0.8545 |
0.2371 | 50.0 | 525 | 0.3307 | 0.8727 |
0.2298 | 50.95 | 535 | 0.2449 | 0.9152 |
0.235 | 52.0 | 546 | 0.3831 | 0.8545 |
0.1972 | 52.95 | 556 | 0.3087 | 0.8424 |
0.1993 | 54.0 | 567 | 0.2912 | 0.8848 |
0.2183 | 54.95 | 577 | 0.3253 | 0.8545 |
0.2222 | 56.0 | 588 | 0.3338 | 0.8727 |
0.1984 | 56.95 | 598 | 0.3510 | 0.8364 |
0.174 | 58.0 | 609 | 0.3521 | 0.8667 |
0.2194 | 58.95 | 619 | 0.2718 | 0.8667 |
0.1734 | 60.0 | 630 | 0.3758 | 0.8667 |
0.1841 | 60.95 | 640 | 0.3342 | 0.8727 |
0.1747 | 62.0 | 651 | 0.3858 | 0.8485 |
0.2196 | 62.95 | 661 | 0.4457 | 0.8121 |
0.1899 | 64.0 | 672 | 0.3924 | 0.8545 |
0.2504 | 64.95 | 682 | 0.3071 | 0.8667 |
0.2099 | 66.0 | 693 | 0.4383 | 0.7879 |
0.1707 | 66.95 | 703 | 0.3140 | 0.8788 |
0.2126 | 68.0 | 714 | 0.3500 | 0.8667 |
0.1703 | 68.95 | 724 | 0.3411 | 0.8606 |
0.1602 | 70.0 | 735 | 0.3394 | 0.8606 |
0.1404 | 70.95 | 745 | 0.3308 | 0.8727 |
0.156 | 72.0 | 756 | 0.3535 | 0.8606 |
0.1305 | 72.95 | 766 | 0.3296 | 0.8606 |
0.1516 | 74.0 | 777 | 0.3859 | 0.8485 |
0.1536 | 74.95 | 787 | 0.3857 | 0.8545 |
0.1434 | 76.0 | 798 | 0.3344 | 0.8667 |
0.1499 | 76.95 | 808 | 0.2926 | 0.8788 |
0.1623 | 78.0 | 819 | 0.3481 | 0.8606 |
0.146 | 78.95 | 829 | 0.3499 | 0.8727 |
0.1457 | 80.0 | 840 | 0.3536 | 0.8909 |
0.1779 | 80.95 | 850 | 0.3358 | 0.8848 |
0.153 | 82.0 | 861 | 0.4687 | 0.8242 |
0.1558 | 82.95 | 871 | 0.3269 | 0.8606 |
0.1594 | 84.0 | 882 | 0.4053 | 0.8545 |
0.1455 | 84.95 | 892 | 0.3744 | 0.8545 |
0.1409 | 86.0 | 903 | 0.2758 | 0.8788 |
0.1364 | 86.95 | 913 | 0.3159 | 0.8788 |
0.1233 | 88.0 | 924 | 0.3728 | 0.8606 |
0.1266 | 88.95 | 934 | 0.4164 | 0.8424 |
0.1239 | 90.0 | 945 | 0.3519 | 0.8848 |
0.1617 | 90.95 | 955 | 0.2978 | 0.8848 |
0.1487 | 92.0 | 966 | 0.2711 | 0.8970 |
0.1045 | 92.95 | 976 | 0.3045 | 0.8788 |
0.1319 | 94.0 | 987 | 0.3578 | 0.8667 |
0.1349 | 94.95 | 997 | 0.2984 | 0.8848 |
0.1053 | 95.24 | 1000 | 0.3329 | 0.8909 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.13.3