All_balanced-mms1ball-adaptertest-Nov28
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: inf
- Wer: 0.4258
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: 1e-05
- 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 |
---|---|---|---|---|
104.7377 | 0.0602 | 100 | inf | 1.0006 |
86.3964 | 0.1204 | 200 | inf | 1.0004 |
71.1469 | 0.1806 | 300 | inf | 1.0 |
63.5388 | 0.2408 | 400 | inf | 1.0 |
36.6288 | 0.3010 | 500 | inf | 1.0 |
24.8799 | 0.3612 | 600 | inf | 1.0 |
13.1157 | 0.4214 | 700 | inf | 1.0 |
6.5013 | 0.4816 | 800 | inf | 1.0 |
4.5713 | 0.5418 | 900 | inf | 1.0 |
4.1285 | 0.6020 | 1000 | inf | 1.0 |
3.7921 | 0.6623 | 1100 | inf | 1.0 |
3.6424 | 0.7225 | 1200 | inf | 1.0 |
3.6716 | 0.7827 | 1300 | inf | 1.0 |
3.5539 | 0.8429 | 1400 | inf | 1.0 |
3.4234 | 0.9031 | 1500 | inf | 1.0 |
3.5372 | 0.9633 | 1600 | inf | 1.0 |
3.4625 | 1.0235 | 1700 | inf | 1.0 |
3.4256 | 1.0837 | 1800 | inf | 1.0 |
3.4454 | 1.1439 | 1900 | inf | 1.0 |
3.4054 | 1.2041 | 2000 | inf | 1.0 |
3.3542 | 1.2643 | 2100 | inf | 1.0 |
3.351 | 1.3245 | 2200 | inf | 1.0 |
3.37 | 1.3847 | 2300 | inf | 1.0 |
3.2994 | 1.4449 | 2400 | inf | 1.0 |
3.3228 | 1.5051 | 2500 | inf | 1.0 |
3.3294 | 1.5653 | 2600 | inf | 1.0 |
3.4365 | 1.6255 | 2700 | inf | 1.0 |
3.2813 | 1.6857 | 2800 | inf | 1.0 |
3.3217 | 1.7459 | 2900 | inf | 1.0 |
3.3397 | 1.8061 | 3000 | inf | 1.0 |
3.2714 | 1.8663 | 3100 | inf | 1.0 |
3.2705 | 1.9266 | 3200 | inf | 1.0 |
3.2597 | 1.9868 | 3300 | inf | 1.0 |
3.2471 | 2.0470 | 3400 | inf | 1.0 |
3.3927 | 2.1072 | 3500 | inf | 1.0 |
3.3093 | 2.1674 | 3600 | inf | 1.0 |
3.2264 | 2.2276 | 3700 | inf | 1.0 |
3.2208 | 2.2878 | 3800 | inf | 1.0 |
3.3372 | 2.3480 | 3900 | inf | 1.0 |
3.1658 | 2.4082 | 4000 | inf | 1.0 |
3.1499 | 2.4684 | 4100 | inf | 1.0 |
3.2204 | 2.5286 | 4200 | inf | 1.0 |
3.2322 | 2.5888 | 4300 | inf | 1.0 |
3.2004 | 2.6490 | 4400 | inf | 1.0 |
3.1913 | 2.7092 | 4500 | inf | 1.0 |
3.1561 | 2.7694 | 4600 | inf | 1.0 |
3.165 | 2.8296 | 4700 | inf | 1.0 |
3.1608 | 2.8898 | 4800 | inf | 1.0 |
3.1087 | 2.9500 | 4900 | inf | 1.0 |
3.159 | 3.0102 | 5000 | inf | 1.0 |
3.0894 | 3.0704 | 5100 | inf | 1.0 |
3.0647 | 3.1306 | 5200 | inf | 1.0 |
3.0851 | 3.1908 | 5300 | inf | 0.9998 |
3.1414 | 3.2511 | 5400 | inf | 0.9996 |
3.1283 | 3.3113 | 5500 | inf | 0.9994 |
3.0102 | 3.3715 | 5600 | inf | 0.9980 |
3.0183 | 3.4317 | 5700 | inf | 0.9982 |
2.9454 | 3.4919 | 5800 | inf | 0.9994 |
2.9309 | 3.5521 | 5900 | inf | 1.0004 |
2.9968 | 3.6123 | 6000 | inf | 1.0045 |
2.9838 | 3.6725 | 6100 | inf | 1.0098 |
2.9452 | 3.7327 | 6200 | inf | 1.0059 |
2.9889 | 3.7929 | 6300 | inf | 1.0012 |
2.8905 | 3.8531 | 6400 | inf | 0.9943 |
2.9325 | 3.9133 | 6500 | inf | 0.9783 |
2.8725 | 3.9735 | 6600 | inf | 0.9604 |
2.7827 | 4.0337 | 6700 | inf | 0.9422 |
2.7264 | 4.0939 | 6800 | inf | 0.9159 |
2.7523 | 4.1541 | 6900 | inf | 0.8921 |
2.8659 | 4.2143 | 7000 | inf | 0.8773 |
2.665 | 4.2745 | 7100 | inf | 0.8486 |
2.8073 | 4.3347 | 7200 | inf | 0.8304 |
2.6979 | 4.3949 | 7300 | inf | 0.8111 |
2.7316 | 4.4551 | 7400 | inf | 0.7860 |
2.6897 | 4.5154 | 7500 | inf | 0.7717 |
2.6828 | 4.5756 | 7600 | inf | 0.7532 |
2.7648 | 4.6358 | 7700 | inf | 0.7438 |
2.6205 | 4.6960 | 7800 | inf | 0.7370 |
2.7043 | 4.7562 | 7900 | inf | 0.7260 |
2.5626 | 4.8164 | 8000 | inf | 0.7085 |
2.5871 | 4.8766 | 8100 | inf | 0.7026 |
2.5814 | 4.9368 | 8200 | inf | 0.6937 |
2.7012 | 4.9970 | 8300 | inf | 0.6825 |
2.4816 | 5.0572 | 8400 | inf | 0.6681 |
2.6789 | 5.1174 | 8500 | inf | 0.6595 |
2.6136 | 5.1776 | 8600 | inf | 0.6519 |
2.5689 | 5.2378 | 8700 | inf | 0.6435 |
2.5708 | 5.2980 | 8800 | inf | 0.6345 |
2.5358 | 5.3582 | 8900 | inf | 0.6287 |
2.4588 | 5.4184 | 9000 | inf | 0.6211 |
2.5932 | 5.4786 | 9100 | inf | 0.6117 |
2.5504 | 5.5388 | 9200 | inf | 0.6010 |
2.4834 | 5.5990 | 9300 | inf | 0.5980 |
2.5113 | 5.6592 | 9400 | inf | 0.5924 |
2.4815 | 5.7194 | 9500 | inf | 0.5883 |
2.5146 | 5.7797 | 9600 | inf | 0.5873 |
2.4457 | 5.8399 | 9700 | inf | 0.5826 |
2.4907 | 5.9001 | 9800 | inf | 0.5774 |
2.5241 | 5.9603 | 9900 | inf | 0.5739 |
2.4866 | 6.0205 | 10000 | inf | 0.5717 |
2.4567 | 6.0807 | 10100 | inf | 0.5684 |
2.3949 | 6.1409 | 10200 | inf | 0.5678 |
2.5562 | 6.2011 | 10300 | inf | 0.5668 |
2.335 | 6.2613 | 10400 | inf | 0.5660 |
2.4115 | 6.3215 | 10500 | inf | 0.5653 |
2.4453 | 6.3817 | 10600 | inf | 0.5641 |
2.4689 | 6.4419 | 10700 | inf | 0.5631 |
2.311 | 6.5021 | 10800 | inf | 0.5612 |
2.3218 | 6.5623 | 10900 | inf | 0.5616 |
2.366 | 6.6225 | 11000 | inf | 0.5575 |
2.3831 | 6.6827 | 11100 | inf | 0.5559 |
2.3986 | 6.7429 | 11200 | inf | 0.5567 |
2.3886 | 6.8031 | 11300 | inf | 0.5540 |
2.3275 | 6.8633 | 11400 | inf | 0.5545 |
2.3585 | 6.9235 | 11500 | inf | 0.5545 |
2.3455 | 6.9837 | 11600 | inf | 0.5526 |
2.4095 | 7.0439 | 11700 | inf | 0.5524 |
2.396 | 7.1042 | 11800 | inf | 0.5510 |
2.3051 | 7.1644 | 11900 | inf | 0.5502 |
2.2471 | 7.2246 | 12000 | inf | 0.5477 |
2.3109 | 7.2848 | 12100 | inf | 0.5458 |
2.4103 | 7.3450 | 12200 | inf | 0.5460 |
2.2821 | 7.4052 | 12300 | inf | 0.5463 |
2.4961 | 7.4654 | 12400 | inf | 0.5461 |
2.1962 | 7.5256 | 12500 | inf | 0.5456 |
2.3257 | 7.5858 | 12600 | inf | 0.5430 |
2.3917 | 7.6460 | 12700 | inf | 0.5419 |
2.2884 | 7.7062 | 12800 | inf | 0.5415 |
2.2477 | 7.7664 | 12900 | inf | 0.5420 |
2.1618 | 7.8266 | 13000 | inf | 0.5405 |
2.2561 | 7.8868 | 13100 | inf | 0.5415 |
2.1719 | 7.9470 | 13200 | inf | 0.5415 |
2.1417 | 8.0072 | 13300 | inf | 0.5389 |
2.2109 | 8.0674 | 13400 | inf | 0.5399 |
2.1978 | 8.1276 | 13500 | inf | 0.5407 |
2.2634 | 8.1878 | 13600 | inf | 0.5397 |
2.2598 | 8.2480 | 13700 | inf | 0.5389 |
2.2579 | 8.3082 | 13800 | inf | 0.5387 |
2.1024 | 8.3685 | 13900 | inf | 0.5378 |
2.2253 | 8.4287 | 14000 | inf | 0.5364 |
2.1518 | 8.4889 | 14100 | inf | 0.5368 |
2.2339 | 8.5491 | 14200 | inf | 0.5344 |
2.2036 | 8.6093 | 14300 | inf | 0.5356 |
2.1635 | 8.6695 | 14400 | inf | 0.5364 |
2.229 | 8.7297 | 14500 | inf | 0.5339 |
2.2803 | 8.7899 | 14600 | inf | 0.5305 |
2.22 | 8.8501 | 14700 | inf | 0.5307 |
2.1981 | 8.9103 | 14800 | inf | 0.5300 |
2.1627 | 8.9705 | 14900 | inf | 0.5321 |
2.1737 | 9.0307 | 15000 | inf | 0.5319 |
2.0855 | 9.0909 | 15100 | inf | 0.5305 |
2.078 | 9.1511 | 15200 | inf | 0.5327 |
2.2162 | 9.2113 | 15300 | inf | 0.5305 |
2.0196 | 9.2715 | 15400 | inf | 0.5298 |
2.1919 | 9.3317 | 15500 | inf | 0.5327 |
2.1344 | 9.3919 | 15600 | inf | 0.5331 |
2.1036 | 9.4521 | 15700 | inf | 0.5327 |
2.2517 | 9.5123 | 15800 | inf | 0.5313 |
2.066 | 9.5725 | 15900 | inf | 0.5296 |
2.1578 | 9.6328 | 16000 | inf | 0.5290 |
2.1122 | 9.6930 | 16100 | inf | 0.5245 |
2.0913 | 9.7532 | 16200 | inf | 0.5239 |
2.0633 | 9.8134 | 16300 | inf | 0.5237 |
2.1646 | 9.8736 | 16400 | inf | 0.5239 |
2.2283 | 9.9338 | 16500 | inf | 0.5223 |
2.1672 | 9.9940 | 16600 | inf | 0.5227 |
2.0768 | 10.0542 | 16700 | inf | 0.5218 |
2.0593 | 10.1144 | 16800 | inf | 0.5212 |
2.0441 | 10.1746 | 16900 | inf | 0.5227 |
2.048 | 10.2348 | 17000 | inf | 0.5190 |
2.2472 | 10.2950 | 17100 | inf | 0.52 |
2.0715 | 10.3552 | 17200 | inf | 0.5177 |
2.1743 | 10.4154 | 17300 | inf | 0.5165 |
2.0812 | 10.4756 | 17400 | inf | 0.5165 |
2.1103 | 10.5358 | 17500 | inf | 0.5149 |
2.1482 | 10.5960 | 17600 | inf | 0.5151 |
2.0297 | 10.6562 | 17700 | inf | 0.5153 |
2.0558 | 10.7164 | 17800 | inf | 0.5147 |
2.0379 | 10.7766 | 17900 | inf | 0.5159 |
1.9573 | 10.8368 | 18000 | inf | 0.5116 |
2.0317 | 10.8970 | 18100 | inf | 0.5116 |
1.9518 | 10.9573 | 18200 | inf | 0.5102 |
1.9623 | 11.0175 | 18300 | inf | 0.5104 |
2.0349 | 11.0777 | 18400 | inf | 0.5069 |
2.0373 | 11.1379 | 18500 | inf | 0.5073 |
2.0961 | 11.1981 | 18600 | inf | 0.5054 |
2.0604 | 11.2583 | 18700 | inf | 0.5065 |
2.0969 | 11.3185 | 18800 | inf | 0.5063 |
2.0262 | 11.3787 | 18900 | inf | 0.5067 |
2.063 | 11.4389 | 19000 | inf | 0.5056 |
2.0065 | 11.4991 | 19100 | inf | 0.5061 |
1.8421 | 11.5593 | 19200 | inf | 0.5056 |
2.0193 | 11.6195 | 19300 | inf | 0.504 |
1.9905 | 11.6797 | 19400 | inf | 0.5050 |
1.8971 | 11.7399 | 19500 | inf | 0.5067 |
1.9896 | 11.8001 | 19600 | inf | 0.5050 |
1.9482 | 11.8603 | 19700 | inf | 0.5063 |
2.0893 | 11.9205 | 19800 | inf | 0.5056 |
2.0149 | 11.9807 | 19900 | inf | 0.5065 |
1.9497 | 12.0409 | 20000 | inf | 0.5044 |
2.0303 | 12.1011 | 20100 | inf | 0.5042 |
2.0212 | 12.1613 | 20200 | inf | 0.504 |
2.0529 | 12.2216 | 20300 | inf | 0.5028 |
1.9422 | 12.2818 | 20400 | inf | 0.5026 |
2.0605 | 12.3420 | 20500 | inf | 0.5032 |
2.0251 | 12.4022 | 20600 | inf | 0.5032 |
1.9428 | 12.4624 | 20700 | inf | 0.5026 |
1.8757 | 12.5226 | 20800 | inf | 0.5020 |
1.9201 | 12.5828 | 20900 | inf | 0.4995 |
2.1063 | 12.6430 | 21000 | inf | 0.4993 |
1.8662 | 12.7032 | 21100 | inf | 0.4985 |
1.9908 | 12.7634 | 21200 | inf | 0.4981 |
1.8985 | 12.8236 | 21300 | inf | 0.4956 |
1.7445 | 12.8838 | 21400 | inf | 0.4964 |
1.8745 | 12.9440 | 21500 | inf | 0.4950 |
1.9364 | 13.0042 | 21600 | inf | 0.4942 |
1.9491 | 13.0644 | 21700 | inf | 0.4939 |
1.9847 | 13.1246 | 21800 | inf | 0.4931 |
1.9001 | 13.1848 | 21900 | inf | 0.4940 |
1.8265 | 13.2450 | 22000 | inf | 0.4923 |
1.8725 | 13.3052 | 22100 | inf | 0.4921 |
2.0441 | 13.3654 | 22200 | inf | 0.4942 |
1.907 | 13.4256 | 22300 | inf | 0.4919 |
1.9421 | 13.4859 | 22400 | inf | 0.4903 |
1.9009 | 13.5461 | 22500 | inf | 0.4886 |
1.7615 | 13.6063 | 22600 | inf | 0.4890 |
1.9819 | 13.6665 | 22700 | inf | 0.4900 |
1.8528 | 13.7267 | 22800 | inf | 0.4927 |
1.9979 | 13.7869 | 22900 | inf | 0.4911 |
1.7754 | 13.8471 | 23000 | inf | 0.4882 |
1.808 | 13.9073 | 23100 | inf | 0.4878 |
1.8841 | 13.9675 | 23200 | inf | 0.4886 |
2.0265 | 14.0277 | 23300 | inf | 0.4847 |
1.7086 | 14.0879 | 23400 | inf | 0.4835 |
1.8442 | 14.1481 | 23500 | inf | 0.4860 |
1.8178 | 14.2083 | 23600 | inf | 0.4864 |
1.866 | 14.2685 | 23700 | inf | 0.4853 |
1.9439 | 14.3287 | 23800 | inf | 0.4847 |
1.7911 | 14.3889 | 23900 | inf | 0.4857 |
1.9575 | 14.4491 | 24000 | inf | 0.4835 |
1.8634 | 14.5093 | 24100 | inf | 0.4843 |
1.8283 | 14.5695 | 24200 | inf | 0.4806 |
1.9249 | 14.6297 | 24300 | inf | 0.4808 |
1.8542 | 14.6899 | 24400 | inf | 0.4788 |
1.842 | 14.7502 | 24500 | inf | 0.4804 |
1.9083 | 14.8104 | 24600 | inf | 0.4794 |
1.901 | 14.8706 | 24700 | inf | 0.4816 |
1.7759 | 14.9308 | 24800 | inf | 0.4814 |
2.0227 | 14.9910 | 24900 | inf | 0.4782 |
1.837 | 15.0512 | 25000 | inf | 0.4790 |
1.8831 | 15.1114 | 25100 | inf | 0.4751 |
1.8904 | 15.1716 | 25200 | inf | 0.4759 |
1.7283 | 15.2318 | 25300 | inf | 0.4753 |
1.7461 | 15.2920 | 25400 | inf | 0.4726 |
1.7337 | 15.3522 | 25500 | inf | 0.4736 |
1.7495 | 15.4124 | 25600 | inf | 0.4699 |
2.0154 | 15.4726 | 25700 | inf | 0.4714 |
1.8853 | 15.5328 | 25800 | inf | 0.4708 |
1.9737 | 15.5930 | 25900 | inf | 0.4681 |
1.7842 | 15.6532 | 26000 | inf | 0.4700 |
1.8163 | 15.7134 | 26100 | inf | 0.4697 |
1.6747 | 15.7736 | 26200 | inf | 0.4722 |
1.7329 | 15.8338 | 26300 | inf | 0.4704 |
1.9267 | 15.8940 | 26400 | inf | 0.4699 |
2.0427 | 15.9542 | 26500 | inf | 0.4685 |
1.7331 | 16.0144 | 26600 | inf | 0.4685 |
1.7991 | 16.0747 | 26700 | inf | 0.4648 |
1.8709 | 16.1349 | 26800 | inf | 0.4628 |
1.7775 | 16.1951 | 26900 | inf | 0.4656 |
1.8327 | 16.2553 | 27000 | inf | 0.4644 |
1.8523 | 16.3155 | 27100 | inf | 0.4650 |
1.8387 | 16.3757 | 27200 | inf | 0.4634 |
1.7927 | 16.4359 | 27300 | inf | 0.4617 |
1.7966 | 16.4961 | 27400 | inf | 0.4609 |
1.8229 | 16.5563 | 27500 | inf | 0.4580 |
1.7343 | 16.6165 | 27600 | inf | 0.4566 |
1.7776 | 16.6767 | 27700 | inf | 0.4570 |
1.7981 | 16.7369 | 27800 | inf | 0.4562 |
1.7284 | 16.7971 | 27900 | inf | 0.4576 |
1.8703 | 16.8573 | 28000 | inf | 0.4574 |
1.6575 | 16.9175 | 28100 | inf | 0.4562 |
1.7891 | 16.9777 | 28200 | inf | 0.4558 |
1.7684 | 17.0379 | 28300 | inf | 0.4537 |
1.8055 | 17.0981 | 28400 | inf | 0.4529 |
1.895 | 17.1583 | 28500 | inf | 0.4519 |
1.7252 | 17.2185 | 28600 | inf | 0.4507 |
1.8919 | 17.2787 | 28700 | inf | 0.4490 |
1.7846 | 17.3390 | 28800 | inf | 0.4503 |
1.8303 | 17.3992 | 28900 | inf | 0.4476 |
1.6867 | 17.4594 | 29000 | inf | 0.4486 |
1.7741 | 17.5196 | 29100 | inf | 0.4496 |
1.9772 | 17.5798 | 29200 | inf | 0.4482 |
1.5981 | 17.6400 | 29300 | inf | 0.4474 |
1.761 | 17.7002 | 29400 | inf | 0.4466 |
1.7124 | 17.7604 | 29500 | inf | 0.4455 |
1.6755 | 17.8206 | 29600 | inf | 0.4468 |
1.7106 | 17.8808 | 29700 | inf | 0.4453 |
1.7005 | 17.9410 | 29800 | inf | 0.4445 |
1.895 | 18.0012 | 29900 | inf | 0.4437 |
1.5989 | 18.0614 | 30000 | inf | 0.4392 |
1.6608 | 18.1216 | 30100 | inf | 0.4406 |
1.7436 | 18.1818 | 30200 | inf | 0.4420 |
1.8082 | 18.2420 | 30300 | inf | 0.4423 |
1.638 | 18.3022 | 30400 | inf | 0.4410 |
1.746 | 18.3624 | 30500 | inf | 0.4402 |
1.7842 | 18.4226 | 30600 | inf | 0.4384 |
1.8155 | 18.4828 | 30700 | inf | 0.4382 |
1.6884 | 18.5430 | 30800 | inf | 0.4365 |
1.9028 | 18.6033 | 30900 | inf | 0.4369 |
1.7206 | 18.6635 | 31000 | inf | 0.4363 |
1.704 | 18.7237 | 31100 | inf | 0.4373 |
1.7998 | 18.7839 | 31200 | inf | 0.4379 |
1.6665 | 18.8441 | 31300 | inf | 0.4375 |
1.9113 | 18.9043 | 31400 | inf | 0.4340 |
1.7787 | 18.9645 | 31500 | inf | 0.4341 |
1.6843 | 19.0247 | 31600 | inf | 0.4330 |
1.6615 | 19.0849 | 31700 | inf | 0.4328 |
1.6965 | 19.1451 | 31800 | inf | 0.4308 |
1.7003 | 19.2053 | 31900 | inf | 0.4302 |
1.7476 | 19.2655 | 32000 | inf | 0.4302 |
1.6831 | 19.3257 | 32100 | inf | 0.4324 |
1.8885 | 19.3859 | 32200 | inf | 0.4287 |
1.6954 | 19.4461 | 32300 | inf | 0.4281 |
1.7848 | 19.5063 | 32400 | inf | 0.4260 |
1.6711 | 19.5665 | 32500 | inf | 0.4265 |
1.6896 | 19.6267 | 32600 | inf | 0.4244 |
1.7142 | 19.6869 | 32700 | inf | 0.4258 |
1.6203 | 19.7471 | 32800 | inf | 0.4252 |
1.6878 | 19.8073 | 32900 | inf | 0.4258 |
Framework versions
- Transformers 4.43.4
- Pytorch 2.4.1
- Datasets 3.0.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for sqrk/All_balanced-mms1ball-adaptertest-Nov28
Base model
facebook/mms-1b-all