<!-- 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-loss-ep10
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: 1.6874
- Hint Hit: 1.6144
- Hint Hit Num: 2.3333
- Hint Precision: 0.4288
- Hint Bleu: 4.2572
- Num: 5.2091
- Gen Len: 18.9881
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Hint Hit | Hint Hit Num | Hint Precision | Hint Bleu | Num | Gen Len |
|---|---|---|---|---|---|---|---|---|---|
| 2.1171 | 0.8 | 200 | 1.7302 | 1.6677 | 2.3728 | 0.4329 | 4.2712 | 5.1934 | 18.9828 |
| 1.8966 | 1.6 | 400 | 1.7124 | 1.5287 | 2.2068 | 0.4093 | 4.0102 | 4.9193 | 18.9899 |
| 1.8335 | 2.4 | 600 | 1.7060 | 1.5822 | 2.2847 | 0.4226 | 4.1554 | 5.0948 | 18.974 |
| 1.7665 | 3.2 | 800 | 1.6994 | 1.6266 | 2.3195 | 0.4258 | 4.226 | 5.0913 | 18.9835 |
| 1.724 | 4.0 | 1000 | 1.6981 | 1.5722 | 2.2723 | 0.4202 | 4.1168 | 5.0725 | 18.9803 |
| 1.6789 | 4.8 | 1200 | 1.6966 | 1.5552 | 2.262 | 0.4209 | 4.1217 | 5.0892 | 18.9803 |
| 1.6399 | 5.6 | 1400 | 1.6874 | 1.6144 | 2.3333 | 0.4288 | 4.2572 | 5.2091 | 18.9881 |
| 1.6234 | 6.4 | 1600 | 1.6956 | 1.5738 | 2.2816 | 0.4205 | 4.157 | 5.1132 | 18.9886 |
| 1.6102 | 7.2 | 1800 | 1.7029 | 1.5715 | 2.2778 | 0.4211 | 4.133 | 5.1031 | 18.9832 |
| 1.5758 | 8.0 | 2000 | 1.7063 | 1.6003 | 2.3076 | 0.425 | 4.1864 | 5.1369 | 18.9807 |
| 1.5654 | 8.8 | 2200 | 1.7129 | 1.5709 | 2.272 | 0.4193 | 4.1171 | 5.0762 | 18.9784 |
| 1.5697 | 9.6 | 2400 | 1.7069 | 1.6118 | 2.3218 | 0.4277 | 4.2164 | 5.162 | 18.9807 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1
