metadata
license: apache-2.0
base_model: sshleifer/distilbart-xsum-12-6
tags:
- generated_from_trainer
model-index:
- name: bart-abs-1509-0313-lr-0.0003-bs-8-maxep-6
results: []
bart-abs-1509-0313-lr-0.0003-bs-8-maxep-6
This model is a fine-tuned version of sshleifer/distilbart-xsum-12-6 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 7.2543
- Rouge/rouge1: 0.3097
- Rouge/rouge2: 0.0856
- Rouge/rougel: 0.2463
- Rouge/rougelsum: 0.2464
- Bertscore/bertscore-precision: 0.8589
- Bertscore/bertscore-recall: 0.8656
- Bertscore/bertscore-f1: 0.8622
- Meteor: 0.2246
- Gen Len: 36.0
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge/rouge1 | Rouge/rouge2 | Rouge/rougel | Rouge/rougelsum | Bertscore/bertscore-precision | Bertscore/bertscore-recall | Bertscore/bertscore-f1 | Meteor | Gen Len |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.3455 | 1.0 | 109 | 4.8900 | 0.3002 | 0.06 | 0.2217 | 0.2214 | 0.8741 | 0.8561 | 0.8649 | 0.2198 | 32.0 |
| 0.7475 | 2.0 | 218 | 5.6171 | 0.2599 | 0.0592 | 0.202 | 0.2017 | 0.8553 | 0.8626 | 0.8589 | 0.2583 | 43.0 |
| 0.5078 | 3.0 | 327 | 6.1951 | 0.2722 | 0.0714 | 0.2029 | 0.2031 | 0.8612 | 0.8618 | 0.8615 | 0.2582 | 44.0 |
| 0.3719 | 4.0 | 436 | 6.6790 | 0.3035 | 0.072 | 0.2428 | 0.2429 | 0.8724 | 0.8571 | 0.8646 | 0.2108 | 29.0 |
| 0.3026 | 5.0 | 545 | 7.0205 | 0.3101 | 0.0691 | 0.2302 | 0.2302 | 0.8569 | 0.8665 | 0.8616 | 0.2291 | 43.0 |
| 0.2546 | 6.0 | 654 | 7.2543 | 0.3097 | 0.0856 | 0.2463 | 0.2464 | 0.8589 | 0.8656 | 0.8622 | 0.2246 | 36.0 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1