<!-- 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. -->
libri-alpha-1-Temp-1-attention-4
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 254.9241
- Wer: 0.2989
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: 2e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1911.3686 | 0.75 | 100 | 495.1192 | 0.5273 |
| 1556.7764 | 1.49 | 200 | 406.3832 | 0.4759 |
| 1347.0447 | 2.24 | 300 | 371.9665 | 0.4413 |
| 1303.7259 | 2.99 | 400 | 350.5383 | 0.4145 |
| 1200.0694 | 3.73 | 500 | 336.2006 | 0.3933 |
| 1177.6688 | 4.48 | 600 | 325.6101 | 0.3803 |
| 1174.224 | 5.22 | 700 | 318.7997 | 0.3697 |
| 1037.0171 | 5.97 | 800 | 310.2211 | 0.3616 |
| 1012.8402 | 6.72 | 900 | 301.0019 | 0.3544 |
| 1012.0772 | 7.46 | 1000 | 295.5239 | 0.3495 |
| 956.3445 | 8.21 | 1100 | 290.5795 | 0.3450 |
| 935.5793 | 8.96 | 1200 | 286.1022 | 0.3408 |
| 955.9658 | 9.7 | 1300 | 281.6091 | 0.3377 |
| 936.7212 | 10.45 | 1400 | 278.7719 | 0.3349 |
| 863.0611 | 11.19 | 1500 | 275.7689 | 0.3268 |
| 871.686 | 11.94 | 1600 | 274.4354 | 0.3296 |
| 891.1862 | 12.69 | 1700 | 271.1785 | 0.3226 |
| 905.8692 | 13.43 | 1800 | 269.1327 | 0.3209 |
| 840.9667 | 14.18 | 1900 | 268.0669 | 0.3160 |
| 856.374 | 14.93 | 2000 | 266.1927 | 0.3174 |
| 853.3527 | 15.67 | 2100 | 265.2582 | 0.3138 |
| 845.1686 | 16.42 | 2200 | 265.0404 | 0.3135 |
| 863.8627 | 17.16 | 2300 | 264.0161 | 0.3127 |
| 812.8962 | 17.91 | 2400 | 262.0590 | 0.3104 |
| 791.2973 | 18.66 | 2500 | 260.7250 | 0.3081 |
| 823.047 | 19.4 | 2600 | 260.4928 | 0.3057 |
| 808.3427 | 20.15 | 2700 | 260.0193 | 0.3066 |
| 787.3638 | 20.9 | 2800 | 258.6302 | 0.3057 |
| 788.9616 | 21.64 | 2900 | 258.6351 | 0.3041 |
| 794.2102 | 22.39 | 3000 | 258.3393 | 0.3027 |
| 805.2969 | 23.13 | 3100 | 257.5751 | 0.3031 |
| 799.9891 | 23.88 | 3200 | 257.0037 | 0.3034 |
| 784.6213 | 24.63 | 3300 | 256.6911 | 0.3024 |
| 809.3116 | 25.37 | 3400 | 255.7936 | 0.3005 |
| 790.4498 | 26.12 | 3500 | 255.7168 | 0.3001 |
| 778.9627 | 26.87 | 3600 | 255.7248 | 0.2995 |
| 754.2412 | 27.61 | 3700 | 255.3348 | 0.3000 |
| 778.785 | 28.36 | 3800 | 254.8843 | 0.2991 |
| 754.3623 | 29.1 | 3900 | 255.0664 | 0.2991 |
| 803.0266 | 29.85 | 4000 | 254.9241 | 0.2989 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1
- Datasets 2.7.0
- Tokenizers 0.11.0
