<!-- 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-cnndm_2000-all-loss-ep20
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: 1.7153
- Hint Hit: 1.7928
- Hint Hit Num: 2.5301
- Hint Precision: 0.4552
- Hint Bleu: 4.636
- Num: 5.479
- Gen Len: 18.9931
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: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Hint Hit | Hint Hit Num | Hint Precision | Hint Bleu | Num | Gen Len |
---|---|---|---|---|---|---|---|---|---|
2.1072 | 1.6 | 200 | 1.7498 | 1.7633 | 2.4767 | 0.4481 | 4.446 | 5.3305 | 18.9731 |
1.8307 | 3.2 | 400 | 1.7198 | 1.6971 | 2.4135 | 0.4417 | 4.4059 | 5.2794 | 18.9886 |
1.7198 | 4.8 | 600 | 1.7166 | 1.7773 | 2.5021 | 0.4525 | 4.551 | 5.4013 | 18.9849 |
1.6269 | 6.4 | 800 | 1.7153 | 1.7928 | 2.5301 | 0.4552 | 4.636 | 5.479 | 18.9931 |
1.566 | 8.0 | 1000 | 1.7258 | 1.7252 | 2.4552 | 0.4471 | 4.495 | 5.3751 | 18.9874 |
1.5007 | 9.6 | 1200 | 1.7465 | 1.6918 | 2.4154 | 0.4414 | 4.4127 | 5.3099 | 18.9844 |
1.4532 | 11.2 | 1400 | 1.7489 | 1.6916 | 2.4277 | 0.4415 | 4.4515 | 5.3719 | 18.9879 |
1.416 | 12.8 | 1600 | 1.7607 | 1.7475 | 2.4828 | 0.4475 | 4.5424 | 5.4239 | 18.988 |
1.3794 | 14.4 | 1800 | 1.7810 | 1.6556 | 2.384 | 0.4342 | 4.3649 | 5.2977 | 18.9847 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1