<!-- 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_4000-all-hint_hit-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: 1.7485
- Hint Hit: 1.114
- Hint Hit Num: 2.074
- Hint Precision: 0.3494
- Hint Bleu: 3.8084
- Num: 5.914
- Gen Len: 19.0
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: 8
- eval_batch_size: 24
- 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 |
---|---|---|---|---|---|---|---|---|---|
2.1456 | 0.4 | 200 | 1.8334 | 0.962 | 1.938 | 0.3358 | 3.6152 | 5.842 | 18.986 |
1.9791 | 0.8 | 400 | 1.7728 | 1.025 | 2.0 | 0.3418 | 3.7262 | 5.9 | 18.986 |
1.8807 | 1.2 | 600 | 1.7601 | 1.007 | 1.942 | 0.3417 | 3.6237 | 5.682 | 18.986 |
1.8082 | 1.6 | 800 | 1.7609 | 1.067 | 2.01 | 0.3468 | 3.6764 | 5.782 | 18.99 |
1.8193 | 2.0 | 1000 | 1.7262 | 1.0745 | 2.02 | 0.3473 | 3.6407 | 5.802 | 19.0 |
1.7098 | 2.4 | 1200 | 1.7295 | 1.087 | 2.016 | 0.3501 | 3.7051 | 5.732 | 18.998 |
1.702 | 2.8 | 1400 | 1.7391 | 0.9745 | 1.936 | 0.3353 | 3.5292 | 5.782 | 19.0 |
1.6659 | 3.2 | 1600 | 1.7512 | 1.0175 | 1.97 | 0.3393 | 3.6971 | 5.78 | 18.996 |
1.6216 | 3.6 | 1800 | 1.7485 | 1.114 | 2.074 | 0.3494 | 3.8084 | 5.914 | 19.0 |
1.6217 | 4.0 | 2000 | 1.7286 | 1.048 | 1.996 | 0.3421 | 3.5672 | 5.788 | 19.0 |
1.5645 | 4.4 | 2200 | 1.7570 | 1.0425 | 2.008 | 0.3415 | 3.6057 | 5.87 | 19.0 |
1.5454 | 4.8 | 2400 | 1.7453 | 1.014 | 1.98 | 0.3375 | 3.5896 | 5.844 | 19.0 |
1.5012 | 5.2 | 2600 | 1.7661 | 1.0905 | 2.048 | 0.3489 | 3.696 | 5.878 | 19.0 |
1.4867 | 5.6 | 2800 | 1.7647 | 1.0795 | 2.034 | 0.3472 | 3.6454 | 5.852 | 19.0 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1