metadata
base_model: google/pegasus-large
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: pegasus-large-finetuned-cnn_dailymail
results: []
pegasus-large-finetuned-cnn_dailymail
This model is a fine-tuned version of google/pegasus-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5155
- Rouge1: 53.2323
- Rouge2: 38.6836
- Rougel: 41.8756
- Rougelsum: 50.7526
- Bleu 1: 39.2946
- Bleu 2: 33.2337
- Bleu 3: 30.2125
- Meteor: 40.4525
- Compression rate: 1.4202
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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Bleu 1 | Bleu 2 | Bleu 3 | Meteor | Compression rate |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.1676 | 1.0 | 5000 | 0.4626 | 58.2243 | 46.8729 | 49.6754 | 56.3425 | 45.8036 | 41.0494 | 38.8167 | 47.5864 | 1.3223 |
| 0.8474 | 2.0 | 10000 | 0.4951 | 51.7813 | 37.5081 | 40.999 | 49.3373 | 38.9702 | 32.8144 | 30.0222 | 39.7542 | 1.3725 |
| 0.7632 | 3.0 | 15000 | 0.4712 | 54.9872 | 41.6279 | 44.557 | 52.6927 | 42.0867 | 36.3443 | 33.5877 | 43.0071 | 1.3649 |
| 0.7009 | 4.0 | 20000 | 0.4875 | 54.5016 | 40.85 | 44.0557 | 52.0705 | 40.2939 | 34.6751 | 31.8994 | 41.8203 | 1.4397 |
| 0.6563 | 5.0 | 25000 | 0.5036 | 52.3997 | 37.6472 | 41.0743 | 49.8349 | 38.2882 | 32.1617 | 29.1582 | 39.4024 | 1.441 |
| 0.6274 | 6.0 | 30000 | 0.5155 | 53.2323 | 38.6836 | 41.8756 | 50.7526 | 39.2946 | 33.2337 | 30.2125 | 40.4525 | 1.4202 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.2+cu118
- Datasets 2.19.0
- Tokenizers 0.19.1