conversation-summ
This model is a fine-tuned version of facebook/bart-large-xsum on the samsum dataset. It achieves the following results on the evaluation set:
- Loss: 0.4048
- Rouge1: 51.7796
- Rouge2: 26.1341
- Rougel: 41.4013
- Rougelsum: 41.4563
- Gen Len: 29.656
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.5781 | 1.0 | 500 | 0.3637 | 50.8871 | 26.6178 | 41.8757 | 41.9291 | 25.16 |
| 0.2183 | 2.0 | 1000 | 0.3586 | 50.7919 | 25.4277 | 40.8428 | 40.8421 | 27.712 |
| 0.1354 | 3.0 | 1500 | 0.4048 | 51.7796 | 26.1341 | 41.4013 | 41.4563 | 29.656 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu116
- Datasets 2.9.0
- Tokenizers 0.13.2