<!-- 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-10-v3
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: 0.5822
- Rouge1: 52.7338
- Rouge2: 30.0274
- Rougel: 48.6696
- Rougelsum: 48.6837
- Meteor: 47.9097
- Bleu-n: 21.1510
- Bleu-1: 52.8292
- Bleu-2: 26.9042
- Bleu-3: 17.0624
- Bleu-4: 11.3263
- Gen Len: 14.3502
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: 0.0001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5958 | 1.0 | 10950 | 0.5637 | 52.0014 | 29.3355 | 48.0977 | 48.1142 | 46.8122 | 20.7193 | 52.9094 | 26.6582 | 17.0250 | 11.4145 | 14.1124 |
| 0.5168 | 2.0 | 21900 | 0.5618 | 52.616 | 29.9824 | 48.6695 | 48.6858 | 47.6756 | 21.2477 | 52.8608 | 26.9682 | 17.2504 | 11.5263 | 14.3338 |
| 0.4748 | 3.0 | 32850 | 0.5682 | 52.6758 | 30.0082 | 48.6759 | 48.6901 | 47.7055 | 21.1081 | 52.9757 | 26.9437 | 17.2084 | 11.4549 | 14.2867 |
| 0.4297 | 4.0 | 43800 | 0.5759 | 52.9434 | 30.2447 | 48.8937 | 48.9019 | 48.0327 | 21.2589 | 53.1581 | 27.1232 | 17.2418 | 11.4865 | 14.3004 |
| 0.4087 | 5.0 | 54750 | 0.5822 | 52.7338 | 30.0274 | 48.6696 | 48.6837 | 47.9097 | 21.1510 | 52.8292 | 26.9042 | 17.0624 | 11.3263 | 14.3502 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
