<!-- 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-base-qg-SQuAD-7
This model is a fine-tuned version of google/flan-t5-base on the squad dataset. It achieves the following results on the evaluation set:
- Loss: 0.5534
- Rouge1: 52.8042
- Rouge2: 30.2086
- Rougel: 48.8966
- Rougelsum: 48.8899
- Meteor: 47.7541
- Bleu-n: 21.4547
- Bleu-1: 53.3739
- Bleu-2: 27.3640
- Bleu-3: 17.6317
- Bleu-4: 11.9291
- Gen Len: 14.2063
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1.429285714e-05
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Meteor | Bleu-n | Bleu-1 | Bleu-2 | Bleu-3 | Bleu-4 | Gen Len |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.6097 | 1.0 | 10950 | 0.5592 | 51.5804 | 28.9366 | 47.7213 | 47.7126 | 46.3686 | 20.4248 | 52.6402 | 26.3750 | 16.8275 | 11.2237 | 14.1062 |
| 0.5614 | 2.0 | 21900 | 0.5536 | 52.3754 | 29.737 | 48.5616 | 48.5539 | 47.2116 | 21.0897 | 53.3324 | 27.1563 | 17.4888 | 11.8018 | 14.1156 |
| 0.547 | 3.0 | 32850 | 0.5530 | 52.621 | 29.9041 | 48.6859 | 48.6805 | 47.4338 | 21.1320 | 53.5817 | 27.3193 | 17.5544 | 11.8181 | 14.1035 |
| 0.5206 | 4.0 | 43800 | 0.5527 | 52.7177 | 30.0935 | 48.8308 | 48.816 | 47.6891 | 21.3821 | 53.4360 | 27.3730 | 17.6143 | 11.8951 | 14.1669 |
| 0.517 | 5.0 | 54750 | 0.5534 | 52.8042 | 30.2086 | 48.8966 | 48.8899 | 47.7541 | 21.4547 | 53.3739 | 27.3640 | 17.6317 | 11.9291 | 14.2063 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
