<!-- 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. -->
retrain_epoch2and3
This model is a fine-tuned version of alexziweiwang/retrain_first1epoch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.4888
- Acc: 0.24
- Wer: 1.0
- Correct: 48
- Total: 200
- Strlen: 200
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: 9e-06
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2.0
Training results
Training Loss | Epoch | Step | Validation Loss | Acc | Wer | Correct | Total | Strlen |
---|---|---|---|---|---|---|---|---|
No log | 0.02 | 5 | 7.8479 | 0.24 | 1.0 | 48 | 200 | 200 |
7.6019 | 0.04 | 10 | 7.4765 | 0.24 | 1.0 | 48 | 200 | 200 |
7.6019 | 0.06 | 15 | 7.1196 | 0.24 | 1.0 | 48 | 200 | 200 |
7.3222 | 0.08 | 20 | 6.8029 | 0.24 | 1.0 | 48 | 200 | 200 |
7.3222 | 0.11 | 25 | 6.5210 | 0.24 | 1.0 | 48 | 200 | 200 |
6.2645 | 0.13 | 30 | 6.2630 | 0.24 | 1.0 | 48 | 200 | 200 |
6.2645 | 0.15 | 35 | 6.0213 | 0.24 | 1.0 | 48 | 200 | 200 |
5.8699 | 0.17 | 40 | 5.8096 | 0.24 | 1.0 | 48 | 200 | 200 |
5.8699 | 0.19 | 45 | 5.5831 | 0.24 | 1.0 | 48 | 200 | 200 |
5.7145 | 0.21 | 50 | 5.3644 | 0.24 | 1.0 | 48 | 200 | 200 |
5.7145 | 0.23 | 55 | 5.1777 | 0.24 | 1.0 | 48 | 200 | 200 |
5.3702 | 0.25 | 60 | 5.0257 | 0.24 | 1.0 | 48 | 200 | 200 |
5.3702 | 0.27 | 65 | 4.8642 | 0.24 | 1.0 | 48 | 200 | 200 |
5.1896 | 0.3 | 70 | 4.7205 | 0.24 | 1.0 | 48 | 200 | 200 |
5.1896 | 0.32 | 75 | 4.5846 | 0.24 | 1.0 | 48 | 200 | 200 |
5.0615 | 0.34 | 80 | 4.4313 | 0.24 | 1.0 | 48 | 200 | 200 |
5.0615 | 0.36 | 85 | 4.2923 | 0.24 | 1.0 | 48 | 200 | 200 |
4.5189 | 0.38 | 90 | 4.1662 | 0.24 | 1.0 | 48 | 200 | 200 |
4.5189 | 0.4 | 95 | 4.0545 | 0.24 | 1.0 | 48 | 200 | 200 |
4.4911 | 0.42 | 100 | 3.9585 | 0.24 | 1.0 | 48 | 200 | 200 |
4.4911 | 0.44 | 105 | 3.8489 | 0.24 | 1.0 | 48 | 200 | 200 |
4.1997 | 0.46 | 110 | 3.7573 | 0.24 | 1.0 | 48 | 200 | 200 |
4.1997 | 0.48 | 115 | 3.6722 | 0.24 | 1.0 | 48 | 200 | 200 |
3.7348 | 0.51 | 120 | 3.5844 | 0.24 | 1.0 | 48 | 200 | 200 |
3.7348 | 0.53 | 125 | 3.4980 | 0.24 | 1.0 | 48 | 200 | 200 |
3.8042 | 0.55 | 130 | 3.4318 | 0.24 | 1.0 | 48 | 200 | 200 |
3.8042 | 0.57 | 135 | 3.3690 | 0.24 | 1.0 | 48 | 200 | 200 |
3.705 | 0.59 | 140 | 3.3126 | 0.24 | 1.0 | 48 | 200 | 200 |
3.705 | 0.61 | 145 | 3.2630 | 0.24 | 1.0 | 48 | 200 | 200 |
3.763 | 0.63 | 150 | 3.2063 | 0.24 | 1.0 | 48 | 200 | 200 |
3.763 | 0.65 | 155 | 3.1562 | 0.24 | 1.0 | 48 | 200 | 200 |
3.5585 | 0.67 | 160 | 3.1096 | 0.24 | 1.0 | 48 | 200 | 200 |
3.5585 | 0.7 | 165 | 3.0719 | 0.24 | 1.0 | 48 | 200 | 200 |
3.213 | 0.72 | 170 | 3.0373 | 0.24 | 1.0 | 48 | 200 | 200 |
3.213 | 0.74 | 175 | 3.0035 | 0.24 | 1.0 | 48 | 200 | 200 |
3.2874 | 0.76 | 180 | 2.9712 | 0.24 | 1.0 | 48 | 200 | 200 |
3.2874 | 0.78 | 185 | 2.9405 | 0.24 | 1.0 | 48 | 200 | 200 |
3.3327 | 0.8 | 190 | 2.9134 | 0.24 | 1.0 | 48 | 200 | 200 |
3.3327 | 0.82 | 195 | 2.8910 | 0.24 | 1.0 | 48 | 200 | 200 |
3.2382 | 0.84 | 200 | 2.8672 | 0.24 | 1.0 | 48 | 200 | 200 |
3.2382 | 0.86 | 205 | 2.8462 | 0.24 | 1.0 | 48 | 200 | 200 |
3.0069 | 0.89 | 210 | 2.8260 | 0.24 | 1.0 | 48 | 200 | 200 |
3.0069 | 0.91 | 215 | 2.8087 | 0.24 | 1.0 | 48 | 200 | 200 |
3.2288 | 0.93 | 220 | 2.7920 | 0.24 | 1.0 | 48 | 200 | 200 |
3.2288 | 0.95 | 225 | 2.7750 | 0.24 | 1.0 | 48 | 200 | 200 |
2.787 | 0.97 | 230 | 2.7557 | 0.24 | 1.0 | 48 | 200 | 200 |
2.787 | 0.99 | 235 | 2.7367 | 0.24 | 1.0 | 48 | 200 | 200 |
2.9717 | 1.01 | 240 | 2.7207 | 0.24 | 1.0 | 48 | 200 | 200 |
2.9717 | 1.03 | 245 | 2.7063 | 0.24 | 1.0 | 48 | 200 | 200 |
2.9269 | 1.05 | 250 | 2.6939 | 0.24 | 1.0 | 48 | 200 | 200 |
2.9269 | 1.08 | 255 | 2.6831 | 0.24 | 1.0 | 48 | 200 | 200 |
2.8771 | 1.1 | 260 | 2.6709 | 0.24 | 1.0 | 48 | 200 | 200 |
2.8771 | 1.12 | 265 | 2.6594 | 0.24 | 1.0 | 48 | 200 | 200 |
3.0474 | 1.14 | 270 | 2.6472 | 0.24 | 1.0 | 48 | 200 | 200 |
3.0474 | 1.16 | 275 | 2.6361 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7652 | 1.18 | 280 | 2.6268 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7652 | 1.2 | 285 | 2.6184 | 0.24 | 1.0 | 48 | 200 | 200 |
2.8322 | 1.22 | 290 | 2.6106 | 0.24 | 1.0 | 48 | 200 | 200 |
2.8322 | 1.24 | 295 | 2.6034 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6464 | 1.27 | 300 | 2.5957 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6464 | 1.29 | 305 | 2.5877 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7974 | 1.31 | 310 | 2.5805 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7974 | 1.33 | 315 | 2.5748 | 0.24 | 1.0 | 48 | 200 | 200 |
2.797 | 1.35 | 320 | 2.5698 | 0.24 | 1.0 | 48 | 200 | 200 |
2.797 | 1.37 | 325 | 2.5644 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7508 | 1.39 | 330 | 2.5595 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7508 | 1.41 | 335 | 2.5537 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7188 | 1.43 | 340 | 2.5486 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7188 | 1.46 | 345 | 2.5434 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6889 | 1.48 | 350 | 2.5377 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6889 | 1.5 | 355 | 2.5336 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6373 | 1.52 | 360 | 2.5300 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6373 | 1.54 | 365 | 2.5258 | 0.24 | 1.0 | 48 | 200 | 200 |
2.765 | 1.56 | 370 | 2.5219 | 0.24 | 1.0 | 48 | 200 | 200 |
2.765 | 1.58 | 375 | 2.5181 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6407 | 1.6 | 380 | 2.5144 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6407 | 1.62 | 385 | 2.5113 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7727 | 1.64 | 390 | 2.5093 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7727 | 1.67 | 395 | 2.5076 | 0.24 | 1.0 | 48 | 200 | 200 |
2.8091 | 1.69 | 400 | 2.5060 | 0.24 | 1.0 | 48 | 200 | 200 |
2.8091 | 1.71 | 405 | 2.5042 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7204 | 1.73 | 410 | 2.5027 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7204 | 1.75 | 415 | 2.5011 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6168 | 1.77 | 420 | 2.4987 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6168 | 1.79 | 425 | 2.4965 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6947 | 1.81 | 430 | 2.4947 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6947 | 1.83 | 435 | 2.4932 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7495 | 1.86 | 440 | 2.4921 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7495 | 1.88 | 445 | 2.4911 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7413 | 1.9 | 450 | 2.4904 | 0.24 | 1.0 | 48 | 200 | 200 |
2.7413 | 1.92 | 455 | 2.4897 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6498 | 1.94 | 460 | 2.4893 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6498 | 1.96 | 465 | 2.4890 | 0.24 | 1.0 | 48 | 200 | 200 |
2.6891 | 1.98 | 470 | 2.4888 | 0.24 | 1.0 | 48 | 200 | 200 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 1.18.3
- Tokenizers 0.13.2