<!-- 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. -->
flan-t5-large-extraction-all-cnn_8000-ep25-nonstop
This model is a fine-tuned version of google/flan-t5-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8671
- Hint Hit Num: 2.008
- Hint Precision: 0.3399
- Num: 5.895
- Gen Len: 18.991
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: 16
- eval_batch_size: 64
- seed: 1799
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | Hint Hit Num | Hint Precision | Num | Gen Len |
|---|---|---|---|---|---|---|---|
| 2.119 | 0.4 | 200 | 1.7746 | 1.966 | 0.3387 | 5.814 | 18.99 |
| 1.9135 | 0.8 | 400 | 1.7157 | 1.76 | 0.3129 | 5.617 | 18.987 |
| 1.85 | 1.2 | 600 | 1.7140 | 1.913 | 0.3327 | 5.732 | 18.995 |
| 1.7963 | 1.6 | 800 | 1.7022 | 1.887 | 0.3291 | 5.702 | 18.994 |
| 1.7784 | 2.0 | 1000 | 1.6911 | 1.875 | 0.3268 | 5.711 | 18.989 |
| 1.711 | 2.4 | 1200 | 1.6935 | 1.932 | 0.3354 | 5.749 | 18.994 |
| 1.7186 | 2.8 | 1400 | 1.6721 | 1.979 | 0.3427 | 5.791 | 18.997 |
| 1.6704 | 3.2 | 1600 | 1.7007 | 1.945 | 0.334 | 5.792 | 18.994 |
| 1.6484 | 3.6 | 1800 | 1.6900 | 1.896 | 0.3282 | 5.751 | 18.994 |
| 1.6334 | 4.0 | 2000 | 1.6732 | 1.879 | 0.3283 | 5.698 | 18.994 |
| 1.5761 | 4.4 | 2200 | 1.6869 | 1.97 | 0.3357 | 5.861 | 18.992 |
| 1.5882 | 4.8 | 2400 | 1.6784 | 1.952 | 0.3354 | 5.792 | 18.992 |
| 1.558 | 5.2 | 2600 | 1.7012 | 1.984 | 0.3394 | 5.83 | 19.0 |
| 1.5339 | 5.6 | 2800 | 1.7013 | 1.898 | 0.3245 | 5.82 | 18.991 |
| 1.5419 | 6.0 | 3000 | 1.6850 | 1.952 | 0.3377 | 5.766 | 18.992 |
| 1.4884 | 6.4 | 3200 | 1.7009 | 1.967 | 0.3375 | 5.812 | 18.991 |
| 1.4857 | 6.8 | 3400 | 1.7038 | 1.913 | 0.3289 | 5.805 | 18.992 |
| 1.4655 | 7.2 | 3600 | 1.7103 | 1.956 | 0.3347 | 5.82 | 18.992 |
| 1.4578 | 7.6 | 3800 | 1.7235 | 1.946 | 0.3318 | 5.837 | 18.999 |
| 1.443 | 8.0 | 4000 | 1.7176 | 1.963 | 0.3347 | 5.828 | 18.991 |
| 1.42 | 8.4 | 4200 | 1.7305 | 1.977 | 0.3404 | 5.809 | 18.996 |
| 1.4155 | 8.8 | 4400 | 1.7267 | 1.988 | 0.3408 | 5.816 | 18.997 |
| 1.3753 | 9.2 | 4600 | 1.7418 | 1.992 | 0.3427 | 5.804 | 19.0 |
| 1.3853 | 9.6 | 4800 | 1.7360 | 2.013 | 0.3461 | 5.818 | 18.992 |
| 1.3768 | 10.0 | 5000 | 1.7280 | 1.994 | 0.3397 | 5.874 | 18.992 |
| 1.3465 | 10.4 | 5200 | 1.7530 | 2.01 | 0.3424 | 5.855 | 18.992 |
| 1.3445 | 10.8 | 5400 | 1.7416 | 1.996 | 0.3438 | 5.814 | 18.992 |
| 1.3321 | 11.2 | 5600 | 1.7653 | 2.014 | 0.3434 | 5.861 | 18.992 |
| 1.3092 | 11.6 | 5800 | 1.7705 | 2.007 | 0.3423 | 5.861 | 18.983 |
| 1.3263 | 12.0 | 6000 | 1.7617 | 1.988 | 0.3412 | 5.815 | 18.986 |
| 1.2847 | 12.4 | 6200 | 1.7816 | 1.988 | 0.3407 | 5.815 | 18.992 |
| 1.2942 | 12.8 | 6400 | 1.7905 | 1.987 | 0.3395 | 5.83 | 18.991 |
| 1.2784 | 13.2 | 6600 | 1.7795 | 2.028 | 0.3436 | 5.899 | 18.992 |
| 1.2562 | 13.6 | 6800 | 1.7861 | 1.97 | 0.3371 | 5.825 | 18.989 |
| 1.2776 | 14.0 | 7000 | 1.7899 | 2.02 | 0.3431 | 5.871 | 18.992 |
| 1.2524 | 14.4 | 7200 | 1.8054 | 2.038 | 0.3435 | 5.916 | 18.992 |
| 1.2402 | 14.8 | 7400 | 1.8072 | 2.034 | 0.3459 | 5.872 | 18.995 |
| 1.2352 | 15.2 | 7600 | 1.8123 | 2.014 | 0.3431 | 5.861 | 18.987 |
| 1.2195 | 15.6 | 7800 | 1.8196 | 2.034 | 0.3444 | 5.869 | 18.987 |
| 1.23 | 16.0 | 8000 | 1.8115 | 1.979 | 0.338 | 5.85 | 18.989 |
| 1.2047 | 16.4 | 8200 | 1.8129 | 2.02 | 0.3428 | 5.888 | 18.99 |
| 1.2155 | 16.8 | 8400 | 1.8178 | 1.978 | 0.335 | 5.883 | 18.991 |
| 1.2028 | 17.2 | 8600 | 1.8293 | 2.017 | 0.3418 | 5.88 | 18.992 |
| 1.189 | 17.6 | 8800 | 1.8303 | 1.983 | 0.3374 | 5.858 | 18.992 |
| 1.195 | 18.0 | 9000 | 1.8367 | 2.021 | 0.3423 | 5.883 | 18.992 |
| 1.1837 | 18.4 | 9200 | 1.8388 | 2.015 | 0.3403 | 5.893 | 18.999 |
| 1.1668 | 18.8 | 9400 | 1.8388 | 2.023 | 0.342 | 5.903 | 18.991 |
| 1.1568 | 19.2 | 9600 | 1.8514 | 2.036 | 0.3458 | 5.876 | 18.99 |
| 1.1783 | 19.6 | 9800 | 1.8419 | 2.042 | 0.3458 | 5.902 | 18.985 |
| 1.1674 | 20.0 | 10000 | 1.8433 | 1.992 | 0.3394 | 5.868 | 18.991 |
| 1.1515 | 20.4 | 10200 | 1.8601 | 2.004 | 0.3404 | 5.881 | 18.985 |
| 1.1478 | 20.8 | 10400 | 1.8520 | 2.032 | 0.3437 | 5.897 | 18.991 |
| 1.1634 | 21.2 | 10600 | 1.8582 | 2.013 | 0.3398 | 5.926 | 18.985 |
| 1.138 | 21.6 | 10800 | 1.8571 | 2.006 | 0.3399 | 5.902 | 18.985 |
| 1.1609 | 22.0 | 11000 | 1.8557 | 2.006 | 0.3402 | 5.899 | 18.991 |
| 1.1306 | 22.4 | 11200 | 1.8622 | 2.02 | 0.3431 | 5.894 | 18.99 |
| 1.1485 | 22.8 | 11400 | 1.8619 | 2.003 | 0.3402 | 5.872 | 18.992 |
| 1.1239 | 23.2 | 11600 | 1.8648 | 2.004 | 0.3405 | 5.879 | 18.992 |
| 1.1427 | 23.6 | 11800 | 1.8651 | 2.003 | 0.3397 | 5.897 | 18.991 |
| 1.1451 | 24.0 | 12000 | 1.8631 | 2.008 | 0.3404 | 5.89 | 18.991 |
| 1.1342 | 24.4 | 12200 | 1.8654 | 2.004 | 0.3397 | 5.884 | 18.99 |
| 1.1289 | 24.8 | 12400 | 1.8672 | 2.005 | 0.3399 | 5.888 | 18.991 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1
