COPAS-mms1ball-Nov30
This model is a fine-tuned version of facebook/mms-1b-all on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3535
- Wer: 0.2197
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: 0.001
- train_batch_size: 3
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 100
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
12.7018 | 0.4405 | 100 | 0.9506 | 0.5515 |
0.7255 | 0.8811 | 200 | 0.6694 | 0.4617 |
0.6192 | 1.3216 | 300 | 0.6035 | 0.4044 |
0.4728 | 1.7621 | 400 | 0.5322 | 0.3760 |
0.4832 | 2.2026 | 500 | 0.5080 | 0.3714 |
0.4645 | 2.6432 | 600 | 0.4722 | 0.3694 |
0.3892 | 3.0837 | 700 | 0.4721 | 0.3765 |
0.3833 | 3.5242 | 800 | 0.4661 | 0.3587 |
0.407 | 3.9648 | 900 | 0.4236 | 0.3506 |
0.3494 | 4.4053 | 1000 | 0.4192 | 0.3349 |
0.3385 | 4.8458 | 1100 | 0.4248 | 0.3409 |
0.3301 | 5.2863 | 1200 | 0.3981 | 0.3161 |
0.3327 | 5.7269 | 1300 | 0.3816 | 0.3044 |
0.3148 | 6.1674 | 1400 | 0.3641 | 0.3095 |
0.297 | 6.6079 | 1500 | 0.4005 | 0.3156 |
0.2834 | 7.0485 | 1600 | 0.3849 | 0.3009 |
0.2719 | 7.4890 | 1700 | 0.3817 | 0.3146 |
0.2738 | 7.9295 | 1800 | 0.3654 | 0.2938 |
0.2305 | 8.3700 | 1900 | 0.3882 | 0.3186 |
0.318 | 8.8106 | 2000 | 0.4165 | 0.3338 |
0.2715 | 9.2511 | 2100 | 0.3609 | 0.2922 |
0.2486 | 9.6916 | 2200 | 0.3728 | 0.3130 |
0.2501 | 10.1322 | 2300 | 0.3586 | 0.3004 |
0.2418 | 10.5727 | 2400 | 0.3571 | 0.2861 |
0.2269 | 11.0132 | 2500 | 0.3488 | 0.2851 |
0.2424 | 11.4537 | 2600 | 0.3913 | 0.3156 |
0.2404 | 11.8943 | 2700 | 0.3889 | 0.3024 |
0.242 | 12.3348 | 2800 | 0.3380 | 0.2608 |
0.2482 | 12.7753 | 2900 | 0.3514 | 0.2760 |
0.2069 | 13.2159 | 3000 | 0.3766 | 0.3044 |
0.2363 | 13.6564 | 3100 | 0.3332 | 0.2740 |
0.2739 | 14.0969 | 3200 | 0.3534 | 0.3283 |
0.2021 | 14.5374 | 3300 | 0.3326 | 0.2821 |
0.1918 | 14.9780 | 3400 | 0.3486 | 0.2770 |
0.2063 | 15.4185 | 3500 | 0.3400 | 0.2887 |
0.1962 | 15.8590 | 3600 | 0.3719 | 0.2770 |
0.2001 | 16.2996 | 3700 | 0.3269 | 0.2775 |
0.1863 | 16.7401 | 3800 | 0.3512 | 0.2953 |
0.2337 | 17.1806 | 3900 | 0.3517 | 0.2816 |
0.1977 | 17.6211 | 4000 | 0.3371 | 0.2760 |
0.203 | 18.0617 | 4100 | 0.3366 | 0.2740 |
0.1587 | 18.5022 | 4200 | 0.3561 | 0.2725 |
0.1903 | 18.9427 | 4300 | 0.3211 | 0.2547 |
0.1927 | 19.3833 | 4400 | 0.3424 | 0.2562 |
0.1863 | 19.8238 | 4500 | 0.3275 | 0.2588 |
0.1491 | 20.2643 | 4600 | 0.3314 | 0.2679 |
0.1834 | 20.7048 | 4700 | 0.3358 | 0.2699 |
0.1675 | 21.1454 | 4800 | 0.3430 | 0.2872 |
0.1611 | 21.5859 | 4900 | 0.3465 | 0.2577 |
0.1783 | 22.0264 | 5000 | 0.3262 | 0.2522 |
0.1577 | 22.4670 | 5100 | 0.3211 | 0.2481 |
0.1493 | 22.9075 | 5200 | 0.3257 | 0.2577 |
0.1716 | 23.3480 | 5300 | 0.3127 | 0.2481 |
0.139 | 23.7885 | 5400 | 0.3178 | 0.2547 |
0.1328 | 24.2291 | 5500 | 0.3390 | 0.2648 |
0.1497 | 24.6696 | 5600 | 0.3630 | 0.2430 |
0.1599 | 25.1101 | 5700 | 0.3570 | 0.2603 |
0.1544 | 25.5507 | 5800 | 0.3517 | 0.2577 |
0.1893 | 25.9912 | 5900 | 0.3302 | 0.2476 |
0.145 | 26.4317 | 6000 | 0.3331 | 0.2653 |
0.1644 | 26.8722 | 6100 | 0.3590 | 0.2867 |
0.1672 | 27.3128 | 6200 | 0.3450 | 0.2572 |
0.1411 | 27.7533 | 6300 | 0.3646 | 0.2501 |
0.1232 | 28.1938 | 6400 | 0.3509 | 0.2501 |
0.1568 | 28.6344 | 6500 | 0.3495 | 0.2623 |
0.139 | 29.0749 | 6600 | 0.3294 | 0.2659 |
0.1358 | 29.5154 | 6700 | 0.3464 | 0.2821 |
0.1876 | 29.9559 | 6800 | 0.3594 | 0.2801 |
0.148 | 30.3965 | 6900 | 0.3453 | 0.2506 |
0.175 | 30.8370 | 7000 | 0.3691 | 0.2760 |
0.1616 | 31.2775 | 7100 | 0.3731 | 0.2618 |
0.1716 | 31.7181 | 7200 | 0.3288 | 0.2582 |
0.1442 | 32.1586 | 7300 | 0.3335 | 0.2598 |
0.1393 | 32.5991 | 7400 | 0.3530 | 0.2588 |
0.1426 | 33.0396 | 7500 | 0.3593 | 0.2831 |
0.1366 | 33.4802 | 7600 | 0.3390 | 0.2364 |
0.1369 | 33.9207 | 7700 | 0.3416 | 0.2486 |
0.1427 | 34.3612 | 7800 | 0.3597 | 0.2532 |
0.1344 | 34.8018 | 7900 | 0.3487 | 0.2516 |
0.1131 | 35.2423 | 8000 | 0.3521 | 0.2511 |
0.1067 | 35.6828 | 8100 | 0.3612 | 0.2390 |
0.1239 | 36.1233 | 8200 | 0.3475 | 0.2420 |
0.1174 | 36.5639 | 8300 | 0.3547 | 0.2577 |
0.1403 | 37.0044 | 8400 | 0.3696 | 0.2562 |
0.1112 | 37.4449 | 8500 | 0.3748 | 0.2588 |
0.1119 | 37.8855 | 8600 | 0.3538 | 0.2486 |
0.1302 | 38.3260 | 8700 | 0.3423 | 0.2430 |
0.1026 | 38.7665 | 8800 | 0.3594 | 0.2476 |
0.0965 | 39.2070 | 8900 | 0.3641 | 0.2577 |
0.0934 | 39.6476 | 9000 | 0.3727 | 0.2613 |
0.1015 | 40.0881 | 9100 | 0.3614 | 0.2491 |
0.1011 | 40.5286 | 9200 | 0.3681 | 0.2506 |
0.0802 | 40.9692 | 9300 | 0.3567 | 0.2633 |
0.0958 | 41.4097 | 9400 | 0.3251 | 0.2354 |
0.0896 | 41.8502 | 9500 | 0.3260 | 0.2461 |
0.0785 | 42.2907 | 9600 | 0.3354 | 0.2385 |
0.0924 | 42.7313 | 9700 | 0.3601 | 0.2481 |
0.0899 | 43.1718 | 9800 | 0.3447 | 0.2369 |
0.0827 | 43.6123 | 9900 | 0.3268 | 0.2354 |
0.0874 | 44.0529 | 10000 | 0.3279 | 0.2405 |
0.0827 | 44.4934 | 10100 | 0.3247 | 0.2390 |
0.0793 | 44.9339 | 10200 | 0.3491 | 0.2390 |
0.0723 | 45.3744 | 10300 | 0.3375 | 0.2293 |
0.0845 | 45.8150 | 10400 | 0.3383 | 0.2369 |
0.0769 | 46.2555 | 10500 | 0.3630 | 0.2501 |
0.073 | 46.6960 | 10600 | 0.3489 | 0.2369 |
0.0814 | 47.1366 | 10700 | 0.3473 | 0.2339 |
0.0828 | 47.5771 | 10800 | 0.3543 | 0.2390 |
0.0676 | 48.0176 | 10900 | 0.3475 | 0.2314 |
0.0726 | 48.4581 | 11000 | 0.3541 | 0.2283 |
0.0824 | 48.8987 | 11100 | 0.3564 | 0.2248 |
0.0818 | 49.3392 | 11200 | 0.3792 | 0.2415 |
0.0694 | 49.7797 | 11300 | 0.3596 | 0.2298 |
0.0664 | 50.2203 | 11400 | 0.3526 | 0.2471 |
0.0726 | 50.6608 | 11500 | 0.3453 | 0.2283 |
0.0765 | 51.1013 | 11600 | 0.3456 | 0.2232 |
0.0575 | 51.5419 | 11700 | 0.3671 | 0.2334 |
0.0661 | 51.9824 | 11800 | 0.3663 | 0.2278 |
0.0764 | 52.4229 | 11900 | 0.3663 | 0.2314 |
0.063 | 52.8634 | 12000 | 0.3665 | 0.2232 |
0.0728 | 53.3040 | 12100 | 0.3448 | 0.2268 |
0.0641 | 53.7445 | 12200 | 0.3552 | 0.2268 |
0.0631 | 54.1850 | 12300 | 0.3641 | 0.2308 |
0.0684 | 54.6256 | 12400 | 0.3606 | 0.2390 |
0.0656 | 55.0661 | 12500 | 0.3531 | 0.2227 |
0.0545 | 55.5066 | 12600 | 0.3515 | 0.2263 |
0.0664 | 55.9471 | 12700 | 0.3712 | 0.2207 |
0.0537 | 56.3877 | 12800 | 0.3540 | 0.2177 |
0.0595 | 56.8282 | 12900 | 0.3596 | 0.2324 |
0.0628 | 57.2687 | 13000 | 0.3641 | 0.2314 |
0.058 | 57.7093 | 13100 | 0.3648 | 0.2253 |
0.0623 | 58.1498 | 13200 | 0.3549 | 0.2329 |
0.0548 | 58.5903 | 13300 | 0.3714 | 0.2237 |
0.0554 | 59.0308 | 13400 | 0.3706 | 0.2258 |
0.0593 | 59.4714 | 13500 | 0.3911 | 0.2293 |
0.0581 | 59.9119 | 13600 | 0.3687 | 0.2334 |
0.0574 | 60.3524 | 13700 | 0.3878 | 0.2314 |
0.066 | 60.7930 | 13800 | 0.3573 | 0.2273 |
0.0493 | 61.2335 | 13900 | 0.3759 | 0.2253 |
0.0529 | 61.6740 | 14000 | 0.3656 | 0.2217 |
0.0543 | 62.1145 | 14100 | 0.3499 | 0.2207 |
0.0517 | 62.5551 | 14200 | 0.3670 | 0.2232 |
0.0512 | 62.9956 | 14300 | 0.3588 | 0.2319 |
0.0534 | 63.4361 | 14400 | 0.3492 | 0.2253 |
0.05 | 63.8767 | 14500 | 0.3543 | 0.2187 |
0.0601 | 64.3172 | 14600 | 0.3642 | 0.2319 |
0.0469 | 64.7577 | 14700 | 0.3733 | 0.2283 |
0.0505 | 65.1982 | 14800 | 0.3677 | 0.2344 |
0.0468 | 65.6388 | 14900 | 0.3648 | 0.2319 |
0.0593 | 66.0793 | 15000 | 0.3767 | 0.2334 |
0.0403 | 66.5198 | 15100 | 0.3548 | 0.2227 |
0.0499 | 66.9604 | 15200 | 0.3680 | 0.2227 |
0.0501 | 67.4009 | 15300 | 0.3588 | 0.2202 |
0.0482 | 67.8414 | 15400 | 0.3558 | 0.2222 |
0.0457 | 68.2819 | 15500 | 0.3700 | 0.2237 |
0.0696 | 68.7225 | 15600 | 0.3655 | 0.2354 |
0.0516 | 69.1630 | 15700 | 0.3608 | 0.2131 |
0.0486 | 69.6035 | 15800 | 0.3568 | 0.2146 |
0.0486 | 70.0441 | 15900 | 0.3515 | 0.2161 |
0.0481 | 70.4846 | 16000 | 0.3562 | 0.2156 |
0.0357 | 70.9251 | 16100 | 0.3610 | 0.2278 |
0.0467 | 71.3656 | 16200 | 0.3738 | 0.2308 |
0.0481 | 71.8062 | 16300 | 0.3790 | 0.2339 |
0.051 | 72.2467 | 16400 | 0.3575 | 0.2253 |
0.0422 | 72.6872 | 16500 | 0.3592 | 0.2248 |
0.0451 | 73.1278 | 16600 | 0.3645 | 0.2263 |
0.0433 | 73.5683 | 16700 | 0.3537 | 0.2146 |
0.0389 | 74.0088 | 16800 | 0.3615 | 0.2131 |
0.0426 | 74.4493 | 16900 | 0.3662 | 0.2106 |
0.0449 | 74.8899 | 17000 | 0.3673 | 0.2192 |
0.0553 | 75.3304 | 17100 | 0.3585 | 0.2116 |
0.0369 | 75.7709 | 17200 | 0.3593 | 0.2065 |
0.0342 | 76.2115 | 17300 | 0.3608 | 0.2222 |
0.0408 | 76.6520 | 17400 | 0.3557 | 0.2131 |
0.0368 | 77.0925 | 17500 | 0.3469 | 0.2171 |
0.0379 | 77.5330 | 17600 | 0.3418 | 0.2141 |
0.0436 | 77.9736 | 17700 | 0.3432 | 0.2126 |
0.0372 | 78.4141 | 17800 | 0.3548 | 0.2222 |
0.0385 | 78.8546 | 17900 | 0.3501 | 0.2146 |
0.0467 | 79.2952 | 18000 | 0.3348 | 0.2161 |
0.0352 | 79.7357 | 18100 | 0.3465 | 0.2141 |
0.0341 | 80.1762 | 18200 | 0.3472 | 0.2156 |
0.0391 | 80.6167 | 18300 | 0.3505 | 0.2182 |
0.0369 | 81.0573 | 18400 | 0.3495 | 0.2171 |
0.0386 | 81.4978 | 18500 | 0.3588 | 0.2166 |
0.0362 | 81.9383 | 18600 | 0.3534 | 0.2197 |
0.0412 | 82.3789 | 18700 | 0.3603 | 0.2171 |
0.0394 | 82.8194 | 18800 | 0.3546 | 0.2182 |
0.0334 | 83.2599 | 18900 | 0.3537 | 0.2151 |
0.039 | 83.7004 | 19000 | 0.3549 | 0.2207 |
0.0301 | 84.1410 | 19100 | 0.3549 | 0.2141 |
0.0389 | 84.5815 | 19200 | 0.3490 | 0.2177 |
0.0361 | 85.0220 | 19300 | 0.3455 | 0.2126 |
0.0329 | 85.4626 | 19400 | 0.3587 | 0.2121 |
0.0363 | 85.9031 | 19500 | 0.3489 | 0.2156 |
0.0388 | 86.3436 | 19600 | 0.3535 | 0.2197 |
0.0342 | 86.7841 | 19700 | 0.3529 | 0.2121 |
0.029 | 87.2247 | 19800 | 0.3519 | 0.2136 |
0.0354 | 87.6652 | 19900 | 0.3417 | 0.2111 |
0.0336 | 88.1057 | 20000 | 0.3444 | 0.2207 |
0.0317 | 88.5463 | 20100 | 0.3452 | 0.2182 |
0.0293 | 88.9868 | 20200 | 0.3520 | 0.2237 |
0.0365 | 89.4273 | 20300 | 0.3514 | 0.2248 |
0.0249 | 89.8678 | 20400 | 0.3521 | 0.2258 |
0.0318 | 90.3084 | 20500 | 0.3547 | 0.2243 |
0.0358 | 90.7489 | 20600 | 0.3473 | 0.2212 |
0.0275 | 91.1894 | 20700 | 0.3521 | 0.2212 |
0.0305 | 91.6300 | 20800 | 0.3535 | 0.2227 |
0.0413 | 92.0705 | 20900 | 0.3601 | 0.2248 |
0.036 | 92.5110 | 21000 | 0.3580 | 0.2253 |
0.0241 | 92.9515 | 21100 | 0.3582 | 0.2187 |
0.0349 | 93.3921 | 21200 | 0.3563 | 0.2207 |
0.0289 | 93.8326 | 21300 | 0.3524 | 0.2136 |
0.0237 | 94.2731 | 21400 | 0.3591 | 0.2212 |
0.0369 | 94.7137 | 21500 | 0.3513 | 0.2197 |
0.0309 | 95.1542 | 21600 | 0.3541 | 0.2192 |
0.0307 | 95.5947 | 21700 | 0.3528 | 0.2243 |
0.029 | 96.0352 | 21800 | 0.3515 | 0.2187 |
0.0292 | 96.4758 | 21900 | 0.3505 | 0.2171 |
0.0298 | 96.9163 | 22000 | 0.3519 | 0.2192 |
0.0299 | 97.3568 | 22100 | 0.3549 | 0.2171 |
0.0284 | 97.7974 | 22200 | 0.3524 | 0.2161 |
0.0268 | 98.2379 | 22300 | 0.3524 | 0.2156 |
0.0325 | 98.6784 | 22400 | 0.3536 | 0.2197 |
0.0295 | 99.1189 | 22500 | 0.3534 | 0.2202 |
0.0233 | 99.5595 | 22600 | 0.3532 | 0.2192 |
0.0269 | 100.0 | 22700 | 0.3535 | 0.2197 |
Framework versions
- Transformers 4.43.4
- Pytorch 2.4.1
- Datasets 3.0.0
- Tokenizers 0.19.1
- Downloads last month
- 9
Model tree for sqrk/COPAS-mms1ball-Nov30
Base model
facebook/mms-1b-all