<!-- 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. -->
translation_flan_base_v4
This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0564
- Bleu: 38.7822
- Gen Len: 5.7727
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: 12
- eval_batch_size: 12
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 34
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 182 | 0.2030 | 38.2159 | 5.8636 |
| No log | 2.0 | 364 | 0.1910 | 38.2159 | 5.9091 |
| 0.342 | 3.0 | 546 | 0.1827 | 38.2159 | 5.8182 |
| 0.342 | 4.0 | 728 | 0.1612 | 38.2159 | 5.8182 |
| 0.342 | 5.0 | 910 | 0.1417 | 38.2159 | 5.8182 |
| 0.2293 | 6.0 | 1092 | 0.1343 | 38.2159 | 5.8182 |
| 0.2293 | 7.0 | 1274 | 0.1076 | 38.3599 | 5.7273 |
| 0.2293 | 8.0 | 1456 | 0.1165 | 38.3599 | 5.7273 |
| 0.172 | 9.0 | 1638 | 0.0947 | 38.3599 | 5.7273 |
| 0.172 | 10.0 | 1820 | 0.0797 | 38.3599 | 5.7727 |
| 0.1443 | 11.0 | 2002 | 0.0751 | 38.5022 | 5.7727 |
| 0.1443 | 12.0 | 2184 | 0.0754 | 38.3599 | 5.7273 |
| 0.1443 | 13.0 | 2366 | 0.0621 | 38.5022 | 5.7727 |
| 0.11 | 14.0 | 2548 | 0.0679 | 38.643 | 5.8182 |
| 0.11 | 15.0 | 2730 | 0.0563 | 38.643 | 5.8182 |
| 0.11 | 16.0 | 2912 | 0.0497 | 38.7822 | 5.7727 |
| 0.095 | 17.0 | 3094 | 0.0560 | 38.7822 | 5.7727 |
| 0.095 | 18.0 | 3276 | 0.0561 | 38.643 | 5.7273 |
| 0.095 | 19.0 | 3458 | 0.0518 | 38.7822 | 5.7727 |
| 0.0986 | 20.0 | 3640 | 0.0524 | 38.7822 | 5.7727 |
| 0.0986 | 21.0 | 3822 | 0.0492 | 38.7822 | 5.7727 |
| 0.0818 | 22.0 | 4004 | 0.0501 | 38.7822 | 5.7727 |
| 0.0818 | 23.0 | 4186 | 0.0541 | 38.7822 | 5.7727 |
| 0.0818 | 24.0 | 4368 | 0.0507 | 38.7822 | 5.7727 |
| 0.0814 | 25.0 | 4550 | 0.0538 | 38.7822 | 5.7727 |
| 0.0814 | 26.0 | 4732 | 0.0546 | 38.7822 | 5.7727 |
| 0.0814 | 27.0 | 4914 | 0.0582 | 38.7822 | 5.7727 |
| 0.0653 | 28.0 | 5096 | 0.0587 | 38.7822 | 5.7727 |
| 0.0653 | 29.0 | 5278 | 0.0583 | 38.7822 | 5.7727 |
| 0.0653 | 30.0 | 5460 | 0.0586 | 38.7822 | 5.7727 |
| 0.0697 | 31.0 | 5642 | 0.0583 | 38.7822 | 5.7727 |
| 0.0697 | 32.0 | 5824 | 0.0569 | 38.7822 | 5.7727 |
| 0.0604 | 33.0 | 6006 | 0.0562 | 38.7822 | 5.7727 |
| 0.0604 | 34.0 | 6188 | 0.0564 | 38.7822 | 5.7727 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
