<!-- 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. -->
225-tiny_tobacco3482_og_simkd
This model is a fine-tuned version of WinKawaks/vit-tiny-patch16-224 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1232.9076
- Accuracy: 0.845
- Brier Loss: 0.2612
- Nll: 1.7130
- F1 Micro: 0.845
- F1 Macro: 0.8331
- Ece: 0.1516
- Aurc: 0.0543
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: 16
- eval_batch_size: 16
- seed: 42
- 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 | Brier Loss | Nll | F1 Micro | F1 Macro | Ece | Aurc |
---|---|---|---|---|---|---|---|---|---|---|
No log | 1.0 | 50 | 1271.0875 | 0.2 | 0.8576 | 4.7335 | 0.2000 | 0.0749 | 0.2387 | 0.5656 |
No log | 2.0 | 100 | 1266.3860 | 0.48 | 0.6981 | 2.5151 | 0.48 | 0.3418 | 0.2977 | 0.2958 |
No log | 3.0 | 150 | 1260.7906 | 0.65 | 0.5606 | 2.2427 | 0.65 | 0.5932 | 0.3175 | 0.1738 |
No log | 4.0 | 200 | 1257.9875 | 0.635 | 0.5237 | 2.6934 | 0.635 | 0.5258 | 0.2661 | 0.1758 |
No log | 5.0 | 250 | 1253.2234 | 0.68 | 0.4767 | 1.7822 | 0.68 | 0.5759 | 0.2604 | 0.1457 |
No log | 6.0 | 300 | 1255.6379 | 0.65 | 0.4880 | 2.1128 | 0.65 | 0.5824 | 0.2301 | 0.1431 |
No log | 7.0 | 350 | 1255.5657 | 0.68 | 0.4868 | 2.1551 | 0.68 | 0.6119 | 0.2091 | 0.1567 |
No log | 8.0 | 400 | 1249.9714 | 0.7 | 0.4528 | 2.3724 | 0.7 | 0.6674 | 0.2404 | 0.1247 |
No log | 9.0 | 450 | 1252.9314 | 0.75 | 0.4210 | 2.0979 | 0.75 | 0.7443 | 0.2484 | 0.1069 |
1335.0542 | 10.0 | 500 | 1259.8634 | 0.685 | 0.4673 | 3.3686 | 0.685 | 0.6315 | 0.2186 | 0.1396 |
1335.0542 | 11.0 | 550 | 1260.2100 | 0.655 | 0.4945 | 3.6990 | 0.655 | 0.5917 | 0.2486 | 0.1514 |
1335.0542 | 12.0 | 600 | 1242.3972 | 0.73 | 0.3849 | 2.4725 | 0.7300 | 0.7323 | 0.2188 | 0.0916 |
1335.0542 | 13.0 | 650 | 1242.8196 | 0.785 | 0.3174 | 2.0359 | 0.785 | 0.7707 | 0.1768 | 0.0635 |
1335.0542 | 14.0 | 700 | 1242.0604 | 0.76 | 0.3600 | 2.5990 | 0.76 | 0.7563 | 0.2073 | 0.0806 |
1335.0542 | 15.0 | 750 | 1249.1082 | 0.81 | 0.3139 | 2.3924 | 0.81 | 0.7738 | 0.2004 | 0.0686 |
1335.0542 | 16.0 | 800 | 1235.6434 | 0.815 | 0.2970 | 2.1356 | 0.815 | 0.7957 | 0.1658 | 0.0874 |
1335.0542 | 17.0 | 850 | 1246.7532 | 0.805 | 0.3122 | 2.3216 | 0.805 | 0.7720 | 0.1863 | 0.0648 |
1335.0542 | 18.0 | 900 | 1245.1010 | 0.8 | 0.3012 | 1.8121 | 0.8000 | 0.7747 | 0.1744 | 0.0587 |
1335.0542 | 19.0 | 950 | 1248.3215 | 0.79 | 0.3271 | 1.9954 | 0.79 | 0.7668 | 0.2001 | 0.0653 |
1322.7586 | 20.0 | 1000 | 1246.8469 | 0.785 | 0.3455 | 2.4830 | 0.785 | 0.7633 | 0.1968 | 0.0788 |
1322.7586 | 21.0 | 1050 | 1248.2886 | 0.8 | 0.3211 | 2.2441 | 0.8000 | 0.7899 | 0.1820 | 0.0692 |
1322.7586 | 22.0 | 1100 | 1243.5780 | 0.805 | 0.3075 | 2.3942 | 0.805 | 0.7995 | 0.1963 | 0.0614 |
1322.7586 | 23.0 | 1150 | 1244.5054 | 0.82 | 0.2993 | 2.0642 | 0.82 | 0.7988 | 0.1923 | 0.0560 |
1322.7586 | 24.0 | 1200 | 1243.2177 | 0.82 | 0.2948 | 1.9610 | 0.82 | 0.8156 | 0.1860 | 0.0573 |
1322.7586 | 25.0 | 1250 | 1244.5066 | 0.8 | 0.3089 | 2.0136 | 0.8000 | 0.7677 | 0.1817 | 0.0511 |
1322.7586 | 26.0 | 1300 | 1241.2290 | 0.835 | 0.2683 | 1.6252 | 0.835 | 0.8179 | 0.1716 | 0.0473 |
1322.7586 | 27.0 | 1350 | 1242.3634 | 0.815 | 0.2971 | 2.1384 | 0.815 | 0.8043 | 0.1805 | 0.0586 |
1322.7586 | 28.0 | 1400 | 1248.5602 | 0.805 | 0.3035 | 2.3228 | 0.805 | 0.7644 | 0.1682 | 0.0628 |
1322.7586 | 29.0 | 1450 | 1241.1305 | 0.825 | 0.2758 | 2.0506 | 0.825 | 0.8003 | 0.1599 | 0.0513 |
1318.3501 | 30.0 | 1500 | 1234.8096 | 0.84 | 0.2547 | 1.8920 | 0.8400 | 0.8217 | 0.1542 | 0.0556 |
1318.3501 | 31.0 | 1550 | 1235.2516 | 0.84 | 0.2426 | 1.8788 | 0.8400 | 0.8250 | 0.1429 | 0.0380 |
1318.3501 | 32.0 | 1600 | 1237.9358 | 0.835 | 0.2643 | 1.7957 | 0.835 | 0.8171 | 0.1596 | 0.0431 |
1318.3501 | 33.0 | 1650 | 1231.1899 | 0.86 | 0.2449 | 1.8820 | 0.8600 | 0.8424 | 0.1565 | 0.0519 |
1318.3501 | 34.0 | 1700 | 1241.4664 | 0.84 | 0.2614 | 1.7047 | 0.8400 | 0.8240 | 0.1771 | 0.0491 |
1318.3501 | 35.0 | 1750 | 1241.1458 | 0.85 | 0.2485 | 1.8466 | 0.85 | 0.8372 | 0.1585 | 0.0403 |
1318.3501 | 36.0 | 1800 | 1238.1477 | 0.845 | 0.2570 | 1.8164 | 0.845 | 0.8236 | 0.1604 | 0.0739 |
1318.3501 | 37.0 | 1850 | 1238.3875 | 0.85 | 0.2646 | 1.9949 | 0.85 | 0.8333 | 0.1638 | 0.0641 |
1318.3501 | 38.0 | 1900 | 1238.3080 | 0.86 | 0.2393 | 1.7820 | 0.8600 | 0.8458 | 0.1528 | 0.0474 |
1318.3501 | 39.0 | 1950 | 1235.3929 | 0.86 | 0.2459 | 1.8287 | 0.8600 | 0.8544 | 0.1636 | 0.0556 |
1315.684 | 40.0 | 2000 | 1239.4463 | 0.86 | 0.2420 | 1.6866 | 0.8600 | 0.8423 | 0.1507 | 0.0378 |
1315.684 | 41.0 | 2050 | 1237.7450 | 0.85 | 0.2523 | 1.9391 | 0.85 | 0.8387 | 0.1452 | 0.0536 |
1315.684 | 42.0 | 2100 | 1237.9618 | 0.86 | 0.2476 | 1.8292 | 0.8600 | 0.8481 | 0.1636 | 0.0509 |
1315.684 | 43.0 | 2150 | 1235.4918 | 0.845 | 0.2661 | 1.9061 | 0.845 | 0.8333 | 0.1551 | 0.0663 |
1315.684 | 44.0 | 2200 | 1239.4510 | 0.865 | 0.2423 | 1.6291 | 0.865 | 0.8515 | 0.1553 | 0.0565 |
1315.684 | 45.0 | 2250 | 1237.6595 | 0.85 | 0.2470 | 1.8245 | 0.85 | 0.8346 | 0.1514 | 0.0554 |
1315.684 | 46.0 | 2300 | 1238.8110 | 0.835 | 0.2543 | 1.8533 | 0.835 | 0.8232 | 0.1356 | 0.0412 |
1315.684 | 47.0 | 2350 | 1240.4524 | 0.855 | 0.2487 | 1.7030 | 0.855 | 0.8489 | 0.1489 | 0.0402 |
1315.684 | 48.0 | 2400 | 1239.2617 | 0.87 | 0.2387 | 1.6849 | 0.87 | 0.8573 | 0.1506 | 0.0404 |
1315.684 | 49.0 | 2450 | 1240.5238 | 0.85 | 0.2544 | 1.8495 | 0.85 | 0.8365 | 0.1514 | 0.0593 |
1313.9472 | 50.0 | 2500 | 1224.2273 | 0.87 | 0.2408 | 1.8714 | 0.87 | 0.8505 | 0.1504 | 0.0757 |
1313.9472 | 51.0 | 2550 | 1239.5197 | 0.85 | 0.2599 | 1.7630 | 0.85 | 0.8371 | 0.1659 | 0.0587 |
1313.9472 | 52.0 | 2600 | 1237.7816 | 0.865 | 0.2353 | 1.7327 | 0.865 | 0.8518 | 0.1456 | 0.0461 |
1313.9472 | 53.0 | 2650 | 1236.0118 | 0.865 | 0.2414 | 1.7887 | 0.865 | 0.8539 | 0.1607 | 0.0614 |
1313.9472 | 54.0 | 2700 | 1236.8806 | 0.875 | 0.2323 | 1.5017 | 0.875 | 0.8657 | 0.1481 | 0.0481 |
1313.9472 | 55.0 | 2750 | 1232.2323 | 0.865 | 0.2215 | 1.6013 | 0.865 | 0.8524 | 0.1395 | 0.0541 |
1313.9472 | 56.0 | 2800 | 1239.0905 | 0.845 | 0.2495 | 1.6543 | 0.845 | 0.8321 | 0.1602 | 0.0401 |
1313.9472 | 57.0 | 2850 | 1232.8143 | 0.86 | 0.2561 | 1.8669 | 0.8600 | 0.8475 | 0.1551 | 0.0573 |
1313.9472 | 58.0 | 2900 | 1239.1449 | 0.86 | 0.2430 | 1.7579 | 0.8600 | 0.8469 | 0.1491 | 0.0413 |
1313.9472 | 59.0 | 2950 | 1233.6656 | 0.86 | 0.2540 | 1.8388 | 0.8600 | 0.8394 | 0.1534 | 0.0723 |
1312.3446 | 60.0 | 3000 | 1237.9213 | 0.86 | 0.2467 | 1.7297 | 0.8600 | 0.8495 | 0.1516 | 0.0413 |
1312.3446 | 61.0 | 3050 | 1234.9781 | 0.855 | 0.2436 | 1.7597 | 0.855 | 0.8445 | 0.1441 | 0.0650 |
1312.3446 | 62.0 | 3100 | 1233.9817 | 0.86 | 0.2396 | 1.7017 | 0.8600 | 0.8446 | 0.1526 | 0.0517 |
1312.3446 | 63.0 | 3150 | 1231.5956 | 0.88 | 0.2261 | 1.8523 | 0.88 | 0.8694 | 0.1428 | 0.0502 |
1312.3446 | 64.0 | 3200 | 1231.5542 | 0.85 | 0.2574 | 1.7764 | 0.85 | 0.8335 | 0.1640 | 0.0445 |
1312.3446 | 65.0 | 3250 | 1235.3212 | 0.86 | 0.2387 | 1.7426 | 0.8600 | 0.8458 | 0.1437 | 0.0569 |
1312.3446 | 66.0 | 3300 | 1234.1420 | 0.86 | 0.2446 | 1.5064 | 0.8600 | 0.8532 | 0.1491 | 0.0504 |
1312.3446 | 67.0 | 3350 | 1234.3502 | 0.855 | 0.2418 | 1.7734 | 0.855 | 0.8426 | 0.1439 | 0.0560 |
1312.3446 | 68.0 | 3400 | 1235.5698 | 0.865 | 0.2367 | 1.6948 | 0.865 | 0.8563 | 0.1520 | 0.0462 |
1312.3446 | 69.0 | 3450 | 1234.6050 | 0.86 | 0.2403 | 1.7511 | 0.8600 | 0.8458 | 0.1405 | 0.0498 |
1311.1923 | 70.0 | 3500 | 1235.3489 | 0.835 | 0.2654 | 1.7495 | 0.835 | 0.8199 | 0.1503 | 0.0512 |
1311.1923 | 71.0 | 3550 | 1233.4445 | 0.855 | 0.2671 | 1.7786 | 0.855 | 0.8465 | 0.1558 | 0.0589 |
1311.1923 | 72.0 | 3600 | 1234.6138 | 0.86 | 0.2543 | 1.6259 | 0.8600 | 0.8487 | 0.1559 | 0.0612 |
1311.1923 | 73.0 | 3650 | 1234.8722 | 0.86 | 0.2407 | 1.5390 | 0.8600 | 0.8487 | 0.1471 | 0.0566 |
1311.1923 | 74.0 | 3700 | 1233.2711 | 0.87 | 0.2436 | 1.7559 | 0.87 | 0.8575 | 0.1554 | 0.0497 |
1311.1923 | 75.0 | 3750 | 1235.2708 | 0.865 | 0.2386 | 1.6956 | 0.865 | 0.8528 | 0.1520 | 0.0554 |
1311.1923 | 76.0 | 3800 | 1233.7223 | 0.865 | 0.2385 | 1.5563 | 0.865 | 0.8511 | 0.1429 | 0.0565 |
1311.1923 | 77.0 | 3850 | 1234.5378 | 0.865 | 0.2441 | 1.5156 | 0.865 | 0.8528 | 0.1633 | 0.0547 |
1311.1923 | 78.0 | 3900 | 1238.3745 | 0.85 | 0.2469 | 1.4876 | 0.85 | 0.8335 | 0.1400 | 0.0465 |
1311.1923 | 79.0 | 3950 | 1237.1874 | 0.86 | 0.2460 | 1.6190 | 0.8600 | 0.8451 | 0.1579 | 0.0537 |
1310.5116 | 80.0 | 4000 | 1235.0160 | 0.865 | 0.2379 | 1.4887 | 0.865 | 0.8618 | 0.1567 | 0.0547 |
1310.5116 | 81.0 | 4050 | 1233.9181 | 0.85 | 0.2590 | 1.7047 | 0.85 | 0.8362 | 0.1631 | 0.0590 |
1310.5116 | 82.0 | 4100 | 1237.2312 | 0.865 | 0.2485 | 1.6650 | 0.865 | 0.8549 | 0.1674 | 0.0462 |
1310.5116 | 83.0 | 4150 | 1234.4546 | 0.86 | 0.2472 | 1.5453 | 0.8600 | 0.8497 | 0.1449 | 0.0578 |
1310.5116 | 84.0 | 4200 | 1230.0541 | 0.85 | 0.2579 | 1.7589 | 0.85 | 0.8371 | 0.1493 | 0.0615 |
1310.5116 | 85.0 | 4250 | 1234.1154 | 0.855 | 0.2523 | 1.6053 | 0.855 | 0.8423 | 0.1560 | 0.0541 |
1310.5116 | 86.0 | 4300 | 1235.0112 | 0.86 | 0.2541 | 1.6794 | 0.8600 | 0.8497 | 0.1582 | 0.0525 |
1310.5116 | 87.0 | 4350 | 1234.1501 | 0.845 | 0.2566 | 1.7223 | 0.845 | 0.8317 | 0.1568 | 0.0496 |
1310.5116 | 88.0 | 4400 | 1233.6084 | 0.85 | 0.2575 | 1.6697 | 0.85 | 0.8401 | 0.1606 | 0.0509 |
1310.5116 | 89.0 | 4450 | 1230.3450 | 0.855 | 0.2541 | 1.6316 | 0.855 | 0.8402 | 0.1397 | 0.0567 |
1309.7242 | 90.0 | 4500 | 1233.0825 | 0.85 | 0.2584 | 1.7262 | 0.85 | 0.8371 | 0.1430 | 0.0509 |
1309.7242 | 91.0 | 4550 | 1235.7081 | 0.855 | 0.2551 | 1.5906 | 0.855 | 0.8402 | 0.1449 | 0.0504 |
1309.7242 | 92.0 | 4600 | 1234.7166 | 0.855 | 0.2545 | 1.6704 | 0.855 | 0.8501 | 0.1431 | 0.0519 |
1309.7242 | 93.0 | 4650 | 1235.1996 | 0.85 | 0.2597 | 1.7262 | 0.85 | 0.8371 | 0.1595 | 0.0525 |
1309.7242 | 94.0 | 4700 | 1233.9705 | 0.845 | 0.2555 | 1.7033 | 0.845 | 0.8331 | 0.1533 | 0.0521 |
1309.7242 | 95.0 | 4750 | 1229.2874 | 0.845 | 0.2605 | 1.7454 | 0.845 | 0.8331 | 0.1535 | 0.0530 |
1309.7242 | 96.0 | 4800 | 1233.8939 | 0.845 | 0.2614 | 1.7398 | 0.845 | 0.8331 | 0.1540 | 0.0527 |
1309.7242 | 97.0 | 4850 | 1234.3517 | 0.85 | 0.2641 | 1.7150 | 0.85 | 0.8371 | 0.1504 | 0.0543 |
1309.7242 | 98.0 | 4900 | 1236.2716 | 0.845 | 0.2594 | 1.6780 | 0.845 | 0.8331 | 0.1440 | 0.0502 |
1309.7242 | 99.0 | 4950 | 1231.1798 | 0.85 | 0.2633 | 1.7550 | 0.85 | 0.8371 | 0.1446 | 0.0565 |
1309.5659 | 100.0 | 5000 | 1232.9076 | 0.845 | 0.2612 | 1.7130 | 0.845 | 0.8331 | 0.1516 | 0.0543 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1.post200
- Datasets 2.9.0
- Tokenizers 0.13.2