<!-- 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. -->
pure-start-epoch2
This model is a fine-tuned version of alexziweiwang/pure-start-epoch1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 7.7447
- 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: 1.0
Training results
Training Loss | Epoch | Step | Validation Loss | Acc | Wer | Correct | Total | Strlen |
---|---|---|---|---|---|---|---|---|
No log | 0.01 | 2 | 20.4002 | 0.095 | 1.0 | 19 | 200 | 200 |
No log | 0.02 | 4 | 19.9080 | 0.095 | 1.0 | 19 | 200 | 200 |
No log | 0.03 | 6 | 19.4711 | 0.095 | 1.0 | 19 | 200 | 200 |
No log | 0.03 | 8 | 19.1535 | 0.095 | 1.0 | 19 | 200 | 200 |
46.6007 | 0.04 | 10 | 18.6684 | 0.095 | 1.0 | 19 | 200 | 200 |
46.6007 | 0.05 | 12 | 18.1640 | 0.095 | 1.0 | 19 | 200 | 200 |
46.6007 | 0.06 | 14 | 17.6937 | 0.095 | 1.0 | 19 | 200 | 200 |
46.6007 | 0.07 | 16 | 17.2710 | 0.095 | 1.0 | 19 | 200 | 200 |
46.6007 | 0.08 | 18 | 16.8469 | 0.095 | 1.0 | 19 | 200 | 200 |
49.1547 | 0.08 | 20 | 16.4418 | 0.095 | 1.0 | 19 | 200 | 200 |
49.1547 | 0.09 | 22 | 16.0409 | 0.095 | 1.0 | 19 | 200 | 200 |
49.1547 | 0.1 | 24 | 15.6677 | 0.095 | 1.0 | 19 | 200 | 200 |
49.1547 | 0.11 | 26 | 15.3291 | 0.095 | 1.0 | 19 | 200 | 200 |
49.1547 | 0.12 | 28 | 15.0097 | 0.095 | 1.0 | 19 | 200 | 200 |
35.1416 | 0.13 | 30 | 14.6776 | 0.095 | 1.0 | 19 | 200 | 200 |
35.1416 | 0.13 | 32 | 14.3788 | 0.095 | 1.0 | 19 | 200 | 200 |
35.1416 | 0.14 | 34 | 14.0924 | 0.095 | 1.0 | 19 | 200 | 200 |
35.1416 | 0.15 | 36 | 13.8133 | 0.095 | 1.0 | 19 | 200 | 200 |
35.1416 | 0.16 | 38 | 13.5539 | 0.095 | 1.0 | 19 | 200 | 200 |
34.4057 | 0.17 | 40 | 13.3095 | 0.095 | 1.0 | 19 | 200 | 200 |
34.4057 | 0.18 | 42 | 13.0804 | 0.095 | 1.0 | 19 | 200 | 200 |
34.4057 | 0.19 | 44 | 12.8580 | 0.105 | 1.0 | 21 | 200 | 200 |
34.4057 | 0.19 | 46 | 12.6532 | 0.115 | 1.0 | 23 | 200 | 200 |
34.4057 | 0.2 | 48 | 12.4532 | 0.13 | 1.0 | 26 | 200 | 200 |
33.2759 | 0.21 | 50 | 12.2452 | 0.14 | 1.0 | 28 | 200 | 200 |
33.2759 | 0.22 | 52 | 12.0666 | 0.13 | 1.0 | 26 | 200 | 200 |
33.2759 | 0.23 | 54 | 11.8976 | 0.165 | 1.0 | 33 | 200 | 200 |
33.2759 | 0.24 | 56 | 11.7373 | 0.175 | 1.0 | 35 | 200 | 200 |
33.2759 | 0.24 | 58 | 11.5933 | 0.17 | 1.0 | 34 | 200 | 200 |
29.8129 | 0.25 | 60 | 11.4281 | 0.15 | 1.0 | 30 | 200 | 200 |
29.8129 | 0.26 | 62 | 11.2665 | 0.14 | 1.0 | 28 | 200 | 200 |
29.8129 | 0.27 | 64 | 11.1158 | 0.145 | 1.0 | 29 | 200 | 200 |
29.8129 | 0.28 | 66 | 10.9840 | 0.135 | 1.0 | 27 | 200 | 200 |
29.8129 | 0.29 | 68 | 10.8502 | 0.15 | 1.0 | 30 | 200 | 200 |
38.792 | 0.3 | 70 | 10.7341 | 0.15 | 1.0 | 30 | 200 | 200 |
38.792 | 0.3 | 72 | 10.6082 | 0.165 | 1.0 | 33 | 200 | 200 |
38.792 | 0.31 | 74 | 10.4944 | 0.18 | 1.0 | 36 | 200 | 200 |
38.792 | 0.32 | 76 | 10.3818 | 0.21 | 1.0 | 42 | 200 | 200 |
38.792 | 0.33 | 78 | 10.2719 | 0.235 | 1.0 | 47 | 200 | 200 |
28.0092 | 0.34 | 80 | 10.1636 | 0.235 | 1.0 | 47 | 200 | 200 |
28.0092 | 0.35 | 82 | 10.0709 | 0.24 | 1.0 | 48 | 200 | 200 |
28.0092 | 0.35 | 84 | 9.9797 | 0.24 | 1.0 | 48 | 200 | 200 |
28.0092 | 0.36 | 86 | 9.8958 | 0.24 | 1.0 | 48 | 200 | 200 |
28.0092 | 0.37 | 88 | 9.7977 | 0.24 | 1.0 | 48 | 200 | 200 |
21.6175 | 0.38 | 90 | 9.7015 | 0.24 | 1.0 | 48 | 200 | 200 |
21.6175 | 0.39 | 92 | 9.6150 | 0.24 | 1.0 | 48 | 200 | 200 |
21.6175 | 0.4 | 94 | 9.5304 | 0.24 | 1.0 | 48 | 200 | 200 |
21.6175 | 0.4 | 96 | 9.4521 | 0.24 | 1.0 | 48 | 200 | 200 |
21.6175 | 0.41 | 98 | 9.3832 | 0.24 | 1.0 | 48 | 200 | 200 |
26.3434 | 0.42 | 100 | 9.3148 | 0.24 | 1.0 | 48 | 200 | 200 |
26.3434 | 0.43 | 102 | 9.2563 | 0.24 | 1.0 | 48 | 200 | 200 |
26.3434 | 0.44 | 104 | 9.1944 | 0.24 | 1.0 | 48 | 200 | 200 |
26.3434 | 0.45 | 106 | 9.1323 | 0.24 | 1.0 | 48 | 200 | 200 |
26.3434 | 0.46 | 108 | 9.0717 | 0.24 | 1.0 | 48 | 200 | 200 |
24.4387 | 0.46 | 110 | 9.0245 | 0.24 | 1.0 | 48 | 200 | 200 |
24.4387 | 0.47 | 112 | 8.9772 | 0.24 | 1.0 | 48 | 200 | 200 |
24.4387 | 0.48 | 114 | 8.9390 | 0.24 | 1.0 | 48 | 200 | 200 |
24.4387 | 0.49 | 116 | 8.9013 | 0.24 | 1.0 | 48 | 200 | 200 |
24.4387 | 0.5 | 118 | 8.8605 | 0.24 | 1.0 | 48 | 200 | 200 |
21.7305 | 0.51 | 120 | 8.8126 | 0.24 | 1.0 | 48 | 200 | 200 |
21.7305 | 0.51 | 122 | 8.7503 | 0.24 | 1.0 | 48 | 200 | 200 |
21.7305 | 0.52 | 124 | 8.6921 | 0.24 | 1.0 | 48 | 200 | 200 |
21.7305 | 0.53 | 126 | 8.6378 | 0.24 | 1.0 | 48 | 200 | 200 |
21.7305 | 0.54 | 128 | 8.5927 | 0.24 | 1.0 | 48 | 200 | 200 |
21.5989 | 0.55 | 130 | 8.5520 | 0.24 | 1.0 | 48 | 200 | 200 |
21.5989 | 0.56 | 132 | 8.5126 | 0.24 | 1.0 | 48 | 200 | 200 |
21.5989 | 0.56 | 134 | 8.4743 | 0.24 | 1.0 | 48 | 200 | 200 |
21.5989 | 0.57 | 136 | 8.4369 | 0.24 | 1.0 | 48 | 200 | 200 |
21.5989 | 0.58 | 138 | 8.3993 | 0.24 | 1.0 | 48 | 200 | 200 |
21.8372 | 0.59 | 140 | 8.3636 | 0.24 | 1.0 | 48 | 200 | 200 |
21.8372 | 0.6 | 142 | 8.3311 | 0.24 | 1.0 | 48 | 200 | 200 |
21.8372 | 0.61 | 144 | 8.2983 | 0.24 | 1.0 | 48 | 200 | 200 |
21.8372 | 0.62 | 146 | 8.2652 | 0.24 | 1.0 | 48 | 200 | 200 |
21.8372 | 0.62 | 148 | 8.2345 | 0.24 | 1.0 | 48 | 200 | 200 |
20.1716 | 0.63 | 150 | 8.2064 | 0.24 | 1.0 | 48 | 200 | 200 |
20.1716 | 0.64 | 152 | 8.1818 | 0.24 | 1.0 | 48 | 200 | 200 |
20.1716 | 0.65 | 154 | 8.1603 | 0.24 | 1.0 | 48 | 200 | 200 |
20.1716 | 0.66 | 156 | 8.1403 | 0.24 | 1.0 | 48 | 200 | 200 |
20.1716 | 0.67 | 158 | 8.1180 | 0.24 | 1.0 | 48 | 200 | 200 |
24.5655 | 0.67 | 160 | 8.0997 | 0.24 | 1.0 | 48 | 200 | 200 |
24.5655 | 0.68 | 162 | 8.0791 | 0.24 | 1.0 | 48 | 200 | 200 |
24.5655 | 0.69 | 164 | 8.0563 | 0.24 | 1.0 | 48 | 200 | 200 |
24.5655 | 0.7 | 166 | 8.0342 | 0.24 | 1.0 | 48 | 200 | 200 |
24.5655 | 0.71 | 168 | 8.0130 | 0.24 | 1.0 | 48 | 200 | 200 |
17.3768 | 0.72 | 170 | 7.9936 | 0.24 | 1.0 | 48 | 200 | 200 |
17.3768 | 0.72 | 172 | 7.9756 | 0.24 | 1.0 | 48 | 200 | 200 |
17.3768 | 0.73 | 174 | 7.9594 | 0.24 | 1.0 | 48 | 200 | 200 |
17.3768 | 0.74 | 176 | 7.9439 | 0.24 | 1.0 | 48 | 200 | 200 |
17.3768 | 0.75 | 178 | 7.9298 | 0.24 | 1.0 | 48 | 200 | 200 |
19.7473 | 0.76 | 180 | 7.9157 | 0.24 | 1.0 | 48 | 200 | 200 |
19.7473 | 0.77 | 182 | 7.9021 | 0.24 | 1.0 | 48 | 200 | 200 |
19.7473 | 0.78 | 184 | 7.8899 | 0.24 | 1.0 | 48 | 200 | 200 |
19.7473 | 0.78 | 186 | 7.8796 | 0.24 | 1.0 | 48 | 200 | 200 |
19.7473 | 0.79 | 188 | 7.8697 | 0.24 | 1.0 | 48 | 200 | 200 |
15.7279 | 0.8 | 190 | 7.8598 | 0.24 | 1.0 | 48 | 200 | 200 |
15.7279 | 0.81 | 192 | 7.8490 | 0.24 | 1.0 | 48 | 200 | 200 |
15.7279 | 0.82 | 194 | 7.8390 | 0.24 | 1.0 | 48 | 200 | 200 |
15.7279 | 0.83 | 196 | 7.8293 | 0.24 | 1.0 | 48 | 200 | 200 |
15.7279 | 0.83 | 198 | 7.8211 | 0.24 | 1.0 | 48 | 200 | 200 |
18.5034 | 0.84 | 200 | 7.8135 | 0.24 | 1.0 | 48 | 200 | 200 |
18.5034 | 0.85 | 202 | 7.8064 | 0.24 | 1.0 | 48 | 200 | 200 |
18.5034 | 0.86 | 204 | 7.7991 | 0.24 | 1.0 | 48 | 200 | 200 |
18.5034 | 0.87 | 206 | 7.7924 | 0.24 | 1.0 | 48 | 200 | 200 |
18.5034 | 0.88 | 208 | 7.7862 | 0.24 | 1.0 | 48 | 200 | 200 |
17.1983 | 0.89 | 210 | 7.7803 | 0.24 | 1.0 | 48 | 200 | 200 |
17.1983 | 0.89 | 212 | 7.7749 | 0.24 | 1.0 | 48 | 200 | 200 |
17.1983 | 0.9 | 214 | 7.7701 | 0.24 | 1.0 | 48 | 200 | 200 |
17.1983 | 0.91 | 216 | 7.7657 | 0.24 | 1.0 | 48 | 200 | 200 |
17.1983 | 0.92 | 218 | 7.7628 | 0.24 | 1.0 | 48 | 200 | 200 |
17.7276 | 0.93 | 220 | 7.7595 | 0.24 | 1.0 | 48 | 200 | 200 |
17.7276 | 0.94 | 222 | 7.7567 | 0.24 | 1.0 | 48 | 200 | 200 |
17.7276 | 0.94 | 224 | 7.7541 | 0.24 | 1.0 | 48 | 200 | 200 |
17.7276 | 0.95 | 226 | 7.7518 | 0.24 | 1.0 | 48 | 200 | 200 |
17.7276 | 0.96 | 228 | 7.7497 | 0.24 | 1.0 | 48 | 200 | 200 |
17.8692 | 0.97 | 230 | 7.7479 | 0.24 | 1.0 | 48 | 200 | 200 |
17.8692 | 0.98 | 232 | 7.7463 | 0.24 | 1.0 | 48 | 200 | 200 |
17.8692 | 0.99 | 234 | 7.7453 | 0.24 | 1.0 | 48 | 200 | 200 |
17.8692 | 0.99 | 236 | 7.7447 | 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