mt5_summnerizing_ru_10_epochs
This model is a fine-tuned version of biakko/mt5_summnerizing_ru_10_epochs on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8222
- Rouge1: 7.0928
- Rouge2: 1.4611
- Rougel: 7.0901
- Rougelsum: 7.1092
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: 5.6e-05
- 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 2.0753 | 1.0 | 4005 | 1.8834 | 7.0715 | 1.4252 | 7.0568 | 7.0873 |
| 2.1141 | 2.0 | 8010 | 1.8454 | 7.1718 | 1.3771 | 7.1736 | 7.1871 |
| 2.1635 | 3.0 | 12015 | 1.8222 | 7.0928 | 1.4611 | 7.0901 | 7.1092 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3