fine_tuned_t5_small_model-naive-firstapproach
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.3883
- Rouge1: 0.3966
- Rouge2: 0.1308
- Rougel: 0.2362
- Rougelsum: 0.236
- Gen Len: 231.6737
- Bert F1: 0.8385
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: 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: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | Bert F1 |
|---|---|---|---|---|---|---|---|---|---|
| 4.6189 | 1.0 | 95 | 3.5459 | 0.3918 | 0.124 | 0.2256 | 0.2254 | 201.7579 | 0.8382 |
| 3.7698 | 2.0 | 190 | 3.4353 | 0.3925 | 0.1292 | 0.2333 | 0.2333 | 221.0842 | 0.8381 |
| 3.6955 | 3.0 | 285 | 3.3999 | 0.3954 | 0.1313 | 0.2352 | 0.2352 | 231.1316 | 0.8382 |
| 3.6645 | 4.0 | 380 | 3.3883 | 0.3966 | 0.1308 | 0.2362 | 0.236 | 231.6737 | 0.8385 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0