<!-- 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_v3
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: 0.0880
- Bleu: 10.6786
- Gen Len: 7.7273
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: 28
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
No log | 1.0 | 182 | 0.1606 | 10.6297 | 7.7273 |
No log | 2.0 | 364 | 0.1432 | 10.6297 | 7.7273 |
0.3575 | 3.0 | 546 | 0.1285 | 10.6297 | 7.7273 |
0.3575 | 4.0 | 728 | 0.1207 | 10.6786 | 7.7273 |
0.3575 | 5.0 | 910 | 0.1107 | 10.6786 | 7.7273 |
0.2194 | 6.0 | 1092 | 0.1013 | 10.6786 | 7.7273 |
0.2194 | 7.0 | 1274 | 0.0943 | 10.6786 | 7.7273 |
0.2194 | 8.0 | 1456 | 0.0873 | 10.6786 | 7.7273 |
0.1839 | 9.0 | 1638 | 0.1019 | 10.6786 | 7.7273 |
0.1839 | 10.0 | 1820 | 0.0935 | 10.6786 | 7.7273 |
0.1336 | 11.0 | 2002 | 0.0880 | 10.6786 | 7.7273 |
0.1336 | 12.0 | 2184 | 0.0813 | 10.6786 | 7.7273 |
0.1336 | 13.0 | 2366 | 0.0831 | 10.6786 | 7.7273 |
0.1142 | 14.0 | 2548 | 0.0789 | 10.6786 | 7.7273 |
0.1142 | 15.0 | 2730 | 0.0797 | 10.6786 | 7.7273 |
0.1142 | 16.0 | 2912 | 0.0784 | 10.6786 | 7.7273 |
0.1071 | 17.0 | 3094 | 0.0858 | 10.6786 | 7.7273 |
0.1071 | 18.0 | 3276 | 0.0862 | 10.6786 | 7.7273 |
0.1071 | 19.0 | 3458 | 0.0823 | 10.6786 | 7.7273 |
0.0967 | 20.0 | 3640 | 0.0840 | 10.6786 | 7.7273 |
0.0967 | 21.0 | 3822 | 0.0813 | 10.6786 | 7.7273 |
0.0919 | 22.0 | 4004 | 0.0874 | 10.6786 | 7.7273 |
0.0919 | 23.0 | 4186 | 0.0877 | 10.6786 | 7.7273 |
0.0919 | 24.0 | 4368 | 0.0877 | 10.6786 | 7.7273 |
0.0841 | 25.0 | 4550 | 0.0870 | 10.6786 | 7.7273 |
0.0841 | 26.0 | 4732 | 0.0878 | 10.6786 | 7.7273 |
0.0841 | 27.0 | 4914 | 0.0878 | 10.6786 | 7.7273 |
0.0805 | 28.0 | 5096 | 0.0880 | 10.6786 | 7.7273 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1