metadata
license: mit
tags:
- generated_from_trainer
metrics:
- rouge
base_model: facebook/bart-large-cnn
model-index:
- name: bart-large-cnn_summarizer_30216
results: []
bart-large-cnn_summarizer_30216
This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9800
- Rouge1: 51.6258
- Rouge2: 33.4629
- Rougel: 40.3034
- Rougelsum: 47.8482
- Gen Len: 105.0622
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: 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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.8653 | 1.0 | 12086 | 0.9274 | 51.1144 | 32.972 | 39.8981 | 47.2905 | 100.2417 |
| 0.6741 | 2.0 | 24172 | 0.9330 | 51.5965 | 33.5021 | 40.4048 | 47.8046 | 103.9732 |
| 0.4802 | 3.0 | 36258 | 0.9800 | 51.6258 | 33.4629 | 40.3034 | 47.8482 | 105.0622 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1