resnet50_fold_4_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: 1.1316
- Accuracy: 0.9647
- F1 Score: 0.9664
- Recall: 0.9692
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 |
|---|---|---|---|---|---|---|
| 2.7876 | 1.0 | 20 | 2.7767 | 0.2340 | 0.1926 | 0.2646 |
| 2.7688 | 2.0 | 40 | 2.7642 | 0.2788 | 0.2392 | 0.2991 |
| 2.7329 | 3.0 | 60 | 2.7404 | 0.3590 | 0.3151 | 0.3345 |
| 2.6881 | 4.0 | 80 | 2.7102 | 0.3878 | 0.2983 | 0.3318 |
| 2.6213 | 5.0 | 100 | 2.6711 | 0.4135 | 0.3370 | 0.3549 |
| 2.5373 | 6.0 | 120 | 2.5738 | 0.4712 | 0.4081 | 0.4215 |
| 2.4059 | 7.0 | 140 | 2.4337 | 0.6058 | 0.5719 | 0.5773 |
| 2.2029 | 8.0 | 160 | 2.2399 | 0.6827 | 0.6608 | 0.6694 |
| 1.9764 | 9.0 | 180 | 1.9648 | 0.7660 | 0.7598 | 0.7581 |
| 1.7622 | 10.0 | 200 | 1.7392 | 0.8109 | 0.8108 | 0.8125 |
| 1.6197 | 11.0 | 220 | 1.5386 | 0.8526 | 0.8520 | 0.8547 |
| 1.4890 | 12.0 | 240 | 1.4154 | 0.9038 | 0.9026 | 0.9063 |
| 1.4343 | 13.0 | 260 | 1.3339 | 0.9135 | 0.9142 | 0.9169 |
| 1.3343 | 14.0 | 280 | 1.2875 | 0.9103 | 0.9113 | 0.9145 |
| 1.3474 | 15.0 | 300 | 1.2767 | 0.9103 | 0.9112 | 0.9157 |
| 1.2912 | 16.0 | 320 | 1.2368 | 0.9263 | 0.9281 | 0.9332 |
| 1.2806 | 17.0 | 340 | 1.2247 | 0.9295 | 0.9309 | 0.9369 |
| 1.2594 | 18.0 | 360 | 1.1994 | 0.9327 | 0.9339 | 0.9394 |
| 1.1737 | 19.0 | 380 | 1.1851 | 0.9359 | 0.9373 | 0.9431 |
| 1.2589 | 20.0 | 400 | 1.1753 | 0.9487 | 0.9504 | 0.9542 |
| 1.1885 | 21.0 | 420 | 1.1794 | 0.9455 | 0.9474 | 0.9518 |
| 1.1790 | 22.0 | 440 | 1.1719 | 0.9519 | 0.9540 | 0.9568 |
| 1.2309 | 23.0 | 460 | 1.1672 | 0.9551 | 0.9575 | 0.9605 |
| 1.1581 | 24.0 | 480 | 1.1660 | 0.9487 | 0.9498 | 0.9524 |
| 1.1820 | 25.0 | 500 | 1.1529 | 0.9551 | 0.9567 | 0.9580 |
| 1.1390 | 26.0 | 520 | 1.1630 | 0.9487 | 0.9498 | 0.9524 |
| 1.1469 | 27.0 | 540 | 1.1615 | 0.9487 | 0.9498 | 0.9524 |
| 1.1267 | 28.0 | 560 | 1.1656 | 0.9487 | 0.9501 | 0.9524 |
| 1.1421 | 29.0 | 580 | 1.1692 | 0.9551 | 0.9566 | 0.9586 |
| 1.1177 | 30.0 | 600 | 1.1556 | 0.9519 | 0.9530 | 0.9549 |
| 1.1025 | 31.0 | 620 | 1.1478 | 0.9551 | 0.9569 | 0.9593 |
| 1.1242 | 32.0 | 640 | 1.1432 | 0.9551 | 0.9570 | 0.9603 |
| 1.0660 | 33.0 | 660 | 1.1552 | 0.9615 | 0.9633 | 0.9655 |
| 1.1102 | 34.0 | 680 | 1.1553 | 0.9583 | 0.9600 | 0.9623 |
| 1.1102 | 35.0 | 700 | 1.1554 | 0.9455 | 0.9465 | 0.9468 |
| 1.0782 | 36.0 | 720 | 1.1303 | 0.9615 | 0.9636 | 0.9661 |
| 1.0878 | 37.0 | 740 | 1.1316 | 0.9647 | 0.9664 | 0.9692 |
| 1.0976 | 38.0 | 760 | 1.1559 | 0.9615 | 0.9632 | 0.9661 |
| 1.0927 | 39.0 | 780 | 1.1404 | 0.9615 | 0.9635 | 0.9667 |
| 1.0691 | 40.0 | 800 | 1.1362 | 0.9583 | 0.9596 | 0.9621 |
| 1.1037 | 41.0 | 820 | 1.1316 | 0.9615 | 0.9629 | 0.9655 |
| 1.0616 | 42.0 | 840 | 1.1195 | 0.9615 | 0.9627 | 0.9642 |
| 1.0838 | 43.0 | 860 | 1.1508 | 0.9615 | 0.9633 | 0.9655 |
| 1.0782 | 44.0 | 880 | 1.1183 | 0.9615 | 0.9624 | 0.9642 |
| 1.0664 | 45.0 | 900 | 1.1364 | 0.9615 | 0.9633 | 0.9655 |
| 1.0756 | 46.0 | 920 | 1.1197 | 0.9583 | 0.9597 | 0.9617 |
| 1.0538 | 47.0 | 940 | 1.1142 | 0.9583 | 0.9592 | 0.9602 |
| 1.0664 | 48.0 | 960 | 1.1452 | 0.9615 | 0.9628 | 0.9652 |
| 1.1020 | 49.0 | 980 | 1.1318 | 0.9583 | 0.9597 | 0.9615 |
| 1.0452 | 50.0 | 1000 | 1.1231 | 0.9583 | 0.9597 | 0.9615 |
| 1.0679 | 51.0 | 1020 | 1.1115 | 0.9647 | 0.9662 | 0.9679 |
| 1.0404 | 52.0 | 1040 | 1.1107 | 0.9615 | 0.9633 | 0.9655 |
| 1.1221 | 53.0 | 1060 | 1.1152 | 0.9615 | 0.9627 | 0.9642 |
| 1.0378 | 54.0 | 1080 | 1.1199 | 0.9551 | 0.9562 | 0.9578 |
| 1.0337 | 55.0 | 1100 | 1.1202 | 0.9551 | 0.9562 | 0.9578 |
| 1.0463 | 56.0 | 1120 | 1.1344 | 0.9583 | 0.9597 | 0.9615 |
| 1.0620 | 57.0 | 1140 | 1.1100 | 0.9615 | 0.9633 | 0.9655 |
| 1.0276 | 58.0 | 1160 | 1.1220 | 0.9583 | 0.9597 | 0.9615 |
| 1.0399 | 59.0 | 1180 | 1.1143 | 0.9583 | 0.9597 | 0.9615 |
| 1.0599 | 60.0 | 1200 | 1.1117 | 0.9615 | 0.9627 | 0.9639 |
| 1.0444 | 61.0 | 1220 | 1.1203 | 0.9583 | 0.9597 | 0.9615 |
| 1.0342 | 62.0 | 1240 | 1.1150 | 0.9583 | 0.9592 | 0.9602 |
| 1.0379 | 63.0 | 1260 | 1.1226 | 0.9551 | 0.9562 | 0.9578 |
| 1.0248 | 64.0 | 1280 | 1.1013 | 0.9647 | 0.9657 | 0.9664 |
| 1.0181 | 65.0 | 1300 | 1.1058 | 0.9615 | 0.9627 | 0.9639 |
| 1.0220 | 66.0 | 1320 | 1.1049 | 0.9583 | 0.9592 | 0.9602 |
| 1.0158 | 67.0 | 1340 | 1.1059 | 0.9615 | 0.9627 | 0.9639 |
| 1.0248 | 68.0 | 1360 | 1.1040 | 0.9551 | 0.9560 | 0.9565 |
| 1.0323 | 69.0 | 1380 | 1.1008 | 0.9615 | 0.9627 | 0.9639 |
| 1.0291 | 70.0 | 1400 | 1.0940 | 0.9647 | 0.9652 | 0.9651 |
| 1.0199 | 71.0 | 1420 | 1.1207 | 0.9551 | 0.9562 | 0.9578 |
| 1.0055 | 72.0 | 1440 | 1.1020 | 0.9583 | 0.9592 | 0.9602 |
| 1.0302 | 73.0 | 1460 | 1.1072 | 0.9583 | 0.9595 | 0.9602 |
| 1.0322 | 74.0 | 1480 | 1.1047 | 0.9551 | 0.9560 | 0.9565 |
| 1.0093 | 75.0 | 1500 | 1.1194 | 0.9583 | 0.9597 | 0.9615 |
| 1.0505 | 76.0 | 1520 | 1.0970 | 0.9615 | 0.9622 | 0.9627 |
| 1.0396 | 77.0 | 1540 | 1.1056 | 0.9551 | 0.9560 | 0.9565 |
| 1.0434 | 78.0 | 1560 | 1.1083 | 0.9615 | 0.9627 | 0.9639 |
| 1.0305 | 79.0 | 1580 | 1.1092 | 0.9615 | 0.9627 | 0.9639 |
| 1.0298 | 80.0 | 1600 | 1.1054 | 0.9615 | 0.9627 | 0.9639 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2