<!-- 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_2000-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: 1.8888
- Hint Hit Num: 2.01
- Hint Precision: 0.3447
- Num: 5.837
- Gen Len: 18.999
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 |
|---|---|---|---|---|---|---|---|
| 2.0904 | 1.6 | 200 | 1.7731 | 1.981 | 0.3417 | 5.823 | 18.993 |
| 1.8113 | 3.2 | 400 | 1.7548 | 1.927 | 0.3363 | 5.741 | 18.995 |
| 1.6986 | 4.8 | 600 | 1.7478 | 1.95 | 0.3349 | 5.825 | 18.995 |
| 1.6024 | 6.4 | 800 | 1.7592 | 1.925 | 0.3317 | 5.808 | 18.994 |
| 1.5406 | 8.0 | 1000 | 1.7628 | 1.97 | 0.339 | 5.815 | 18.995 |
| 1.4624 | 9.6 | 1200 | 1.7749 | 1.974 | 0.3378 | 5.839 | 18.988 |
| 1.4123 | 11.2 | 1400 | 1.7986 | 1.952 | 0.3355 | 5.82 | 18.995 |
| 1.3698 | 12.8 | 1600 | 1.8116 | 1.99 | 0.3413 | 5.841 | 18.999 |
| 1.3246 | 14.4 | 1800 | 1.8241 | 1.97 | 0.3368 | 5.842 | 18.999 |
| 1.2865 | 16.0 | 2000 | 1.8407 | 2.016 | 0.3467 | 5.827 | 18.999 |
| 1.2549 | 17.6 | 2200 | 1.8636 | 2.023 | 0.3427 | 5.905 | 18.999 |
| 1.2345 | 19.2 | 2400 | 1.8752 | 2.018 | 0.3439 | 5.874 | 18.999 |
| 1.2122 | 20.8 | 2600 | 1.8750 | 2.001 | 0.3438 | 5.827 | 18.999 |
| 1.2002 | 22.4 | 2800 | 1.8814 | 2.026 | 0.3461 | 5.868 | 18.999 |
| 1.1957 | 24.0 | 3000 | 1.8862 | 2.015 | 0.3457 | 5.832 | 18.999 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1
