pegasus-newsroom-cnn-adam8bit-bs4x64acc_2
This model is a fine-tuned version of oMateos2020/pegasus-newsroom-cnn-adam8bit-bs16x64acc on the cnn_dailymail dataset. It achieves the following results on the evaluation set:
- Loss: 2.8608
- Rouge1: 44.2848
- Rouge2: 21.5452
- Rougel: 31.3765
- Rougelsum: 41.2302
- Gen Len: 71.7744
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: 6.4e-05
- train_batch_size: 4
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.4
- num_epochs: 1
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.9307 | 1.0 | 1121 | 2.8608 | 44.2848 | 21.5452 | 31.3765 | 41.2302 | 71.7744 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1