ml_sum_v2
This model is a fine-tuned version of mHossain/ml_sum_v1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.9401
- Rouge1: 8.1448
- Rouge2: 3.3615
- Rougel: 7.4641
- Rougelsum: 7.9361
- Gen Len: 19.0
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5000
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 312 | 2.1706 | 7.2919 | 2.8117 | 6.7418 | 7.1173 | 19.0 |
| 2.4911 | 2.0 | 625 | 2.1012 | 7.7986 | 3.0952 | 7.1505 | 7.5818 | 19.0 |
| 2.4911 | 3.0 | 937 | 2.0373 | 8.0535 | 3.2228 | 7.3877 | 7.8365 | 19.0 |
| 2.3572 | 4.0 | 1250 | 1.9865 | 8.1591 | 3.31 | 7.4577 | 7.9114 | 19.0 |
| 2.2455 | 4.99 | 1560 | 1.9401 | 8.1448 | 3.3615 | 7.4641 | 7.9361 | 19.0 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2