resnet50_fold_1_v3
This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.1580
- Accuracy: 0.9647
- F1 Score: 0.9654
- Recall: 0.9673
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 150
- num_epochs: 100
- label_smoothing_factor: 0.15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall |
|---|---|---|---|---|---|---|
| 2.7557 | 1.0 | 20 | 2.7625 | 0.2692 | 0.2594 | 0.2612 |
| 2.7387 | 2.0 | 40 | 2.7482 | 0.3141 | 0.2617 | 0.2847 |
| 2.6979 | 3.0 | 60 | 2.7262 | 0.3878 | 0.2740 | 0.3281 |
| 2.6486 | 4.0 | 80 | 2.6881 | 0.4231 | 0.2920 | 0.3538 |
| 2.5751 | 5.0 | 100 | 2.6298 | 0.5064 | 0.4160 | 0.4486 |
| 2.4857 | 6.0 | 120 | 2.5060 | 0.5994 | 0.5319 | 0.5598 |
| 2.3204 | 7.0 | 140 | 2.3467 | 0.6795 | 0.6336 | 0.6488 |
| 2.1366 | 8.0 | 160 | 2.1263 | 0.7532 | 0.7320 | 0.7322 |
| 1.8969 | 9.0 | 180 | 1.8718 | 0.8077 | 0.7996 | 0.7961 |
| 1.7379 | 10.0 | 200 | 1.6131 | 0.8590 | 0.8563 | 0.8539 |
| 1.5923 | 11.0 | 220 | 1.4873 | 0.8846 | 0.8826 | 0.8816 |
| 1.3941 | 12.0 | 240 | 1.3840 | 0.8942 | 0.8898 | 0.8890 |
| 1.3773 | 13.0 | 260 | 1.3154 | 0.9038 | 0.8982 | 0.8958 |
| 1.3183 | 14.0 | 280 | 1.2813 | 0.9199 | 0.9171 | 0.9182 |
| 1.2701 | 15.0 | 300 | 1.2505 | 0.9327 | 0.9301 | 0.9294 |
| 1.2894 | 16.0 | 320 | 1.2321 | 0.9327 | 0.9314 | 0.9300 |
| 1.2239 | 17.0 | 340 | 1.2421 | 0.9263 | 0.9244 | 0.9235 |
| 1.2053 | 18.0 | 360 | 1.2268 | 0.9295 | 0.9289 | 0.9284 |
| 1.1995 | 19.0 | 380 | 1.2101 | 0.9391 | 0.9385 | 0.9386 |
| 1.2248 | 20.0 | 400 | 1.2013 | 0.9391 | 0.9372 | 0.9346 |
| 1.1555 | 21.0 | 420 | 1.1986 | 0.9359 | 0.9352 | 0.9349 |
| 1.1670 | 22.0 | 440 | 1.1890 | 0.9455 | 0.9456 | 0.9453 |
| 1.2071 | 23.0 | 460 | 1.1926 | 0.9423 | 0.9418 | 0.9434 |
| 1.1267 | 24.0 | 480 | 1.2246 | 0.9391 | 0.9383 | 0.9409 |
| 1.1614 | 25.0 | 500 | 1.1970 | 0.9455 | 0.9458 | 0.9465 |
| 1.1558 | 26.0 | 520 | 1.2032 | 0.9423 | 0.9418 | 0.9428 |
| 1.1441 | 27.0 | 540 | 1.1938 | 0.9487 | 0.9488 | 0.9502 |
| 1.1256 | 28.0 | 560 | 1.1955 | 0.9423 | 0.9426 | 0.9441 |
| 1.0955 | 29.0 | 580 | 1.2166 | 0.9327 | 0.9307 | 0.9311 |
| 1.0886 | 30.0 | 600 | 1.2201 | 0.9199 | 0.9176 | 0.9183 |
| 1.1150 | 31.0 | 620 | 1.1919 | 0.9423 | 0.9427 | 0.9441 |
| 1.1152 | 32.0 | 640 | 1.1871 | 0.9391 | 0.9393 | 0.9379 |
| 1.0707 | 33.0 | 660 | 1.2011 | 0.9327 | 0.9330 | 0.9318 |
| 1.0944 | 34.0 | 680 | 1.1869 | 0.9423 | 0.9425 | 0.9416 |
| 1.1264 | 35.0 | 700 | 1.1817 | 0.9487 | 0.9485 | 0.9477 |
| 1.0992 | 36.0 | 720 | 1.1838 | 0.9423 | 0.9425 | 0.9416 |
| 1.0791 | 37.0 | 740 | 1.1749 | 0.9455 | 0.9458 | 0.9453 |
| 1.0995 | 38.0 | 760 | 1.1811 | 0.9551 | 0.9559 | 0.9575 |
| 1.0542 | 39.0 | 780 | 1.1732 | 0.9519 | 0.9522 | 0.9514 |
| 1.1183 | 40.0 | 800 | 1.1655 | 0.9583 | 0.9590 | 0.9612 |
| 1.0785 | 41.0 | 820 | 1.1727 | 0.9455 | 0.9457 | 0.9465 |
| 1.0837 | 42.0 | 840 | 1.1645 | 0.9455 | 0.9458 | 0.9465 |
| 1.0621 | 43.0 | 860 | 1.1836 | 0.9487 | 0.9491 | 0.9514 |
| 1.0465 | 44.0 | 880 | 1.1845 | 0.9423 | 0.9426 | 0.9428 |
| 1.0469 | 45.0 | 900 | 1.1702 | 0.9455 | 0.9461 | 0.9453 |
| 1.0297 | 46.0 | 920 | 1.1686 | 0.9551 | 0.9554 | 0.9575 |
| 1.0551 | 47.0 | 940 | 1.1659 | 0.9583 | 0.9585 | 0.9612 |
| 1.0465 | 48.0 | 960 | 1.1744 | 0.9519 | 0.9526 | 0.9563 |
| 1.0796 | 49.0 | 980 | 1.1688 | 0.9519 | 0.9522 | 0.9539 |
| 1.0476 | 50.0 | 1000 | 1.1593 | 0.9551 | 0.9552 | 0.9563 |
| 1.0353 | 51.0 | 1020 | 1.1657 | 0.9487 | 0.9486 | 0.9477 |
| 1.0581 | 52.0 | 1040 | 1.1620 | 0.9519 | 0.9524 | 0.9514 |
| 1.1047 | 53.0 | 1060 | 1.1604 | 0.9551 | 0.9550 | 0.9551 |
| 1.0317 | 54.0 | 1080 | 1.1661 | 0.9455 | 0.9452 | 0.9441 |
| 1.0229 | 55.0 | 1100 | 1.1587 | 0.9487 | 0.9492 | 0.9477 |
| 1.0533 | 56.0 | 1120 | 1.1624 | 0.9551 | 0.9552 | 0.9563 |
| 1.0438 | 57.0 | 1140 | 1.1634 | 0.9551 | 0.9552 | 0.9563 |
| 1.0314 | 58.0 | 1160 | 1.1543 | 0.9551 | 0.9552 | 0.9563 |
| 1.0473 | 59.0 | 1180 | 1.1591 | 0.9551 | 0.9550 | 0.9551 |
| 1.0623 | 60.0 | 1200 | 1.1580 | 0.9647 | 0.9654 | 0.9673 |
| 1.0197 | 61.0 | 1220 | 1.1587 | 0.9615 | 0.9615 | 0.9637 |
| 1.0110 | 62.0 | 1240 | 1.1541 | 0.9551 | 0.9552 | 0.9563 |
| 1.0129 | 63.0 | 1260 | 1.1455 | 0.9615 | 0.9623 | 0.9637 |
| 1.0515 | 64.0 | 1280 | 1.1501 | 0.9583 | 0.9592 | 0.9600 |
| 1.0202 | 65.0 | 1300 | 1.1541 | 0.9519 | 0.9520 | 0.9526 |
| 1.0114 | 66.0 | 1320 | 1.1557 | 0.9583 | 0.9592 | 0.9600 |
| 1.0040 | 67.0 | 1340 | 1.1573 | 0.9551 | 0.9552 | 0.9563 |
| 1.0181 | 68.0 | 1360 | 1.1559 | 0.9551 | 0.9552 | 0.9563 |
| 1.0284 | 69.0 | 1380 | 1.1523 | 0.9551 | 0.9552 | 0.9563 |
| 1.0234 | 70.0 | 1400 | 1.1595 | 0.9551 | 0.9552 | 0.9563 |
| 1.0203 | 71.0 | 1420 | 1.1545 | 0.9551 | 0.9554 | 0.9575 |
| 1.0216 | 72.0 | 1440 | 1.1493 | 0.9551 | 0.9550 | 0.9551 |
| 1.0271 | 73.0 | 1460 | 1.1642 | 0.9551 | 0.9560 | 0.9563 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2