metadata
license: apache-2.0
base_model: sshleifer/distilbart-xsum-12-6
tags:
- generated_from_trainer
model-index:
- name: bart-abs-1509-0313-lr-3e-05-bs-4-maxep-6
results: []
bart-abs-1509-0313-lr-3e-05-bs-4-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: 2.3615
- Rouge/rouge1: 0.4759
- Rouge/rouge2: 0.2278
- Rouge/rougel: 0.407
- Rouge/rougelsum: 0.4084
- Bertscore/bertscore-precision: 0.898
- Bertscore/bertscore-recall: 0.8937
- Bertscore/bertscore-f1: 0.8957
- Meteor: 0.4311
- Gen Len: 37.0
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2.3606 | 1.0 | 217 | 2.0792 | 0.44 | 0.2028 | 0.3696 | 0.3707 | 0.8966 | 0.8883 | 0.8923 | 0.3866 | 37.3636 |
| 1.6544 | 2.0 | 434 | 1.9952 | 0.4521 | 0.2171 | 0.3902 | 0.3909 | 0.8991 | 0.8912 | 0.895 | 0.4007 | 34.6636 |
| 1.2907 | 3.0 | 651 | 2.0614 | 0.4661 | 0.2212 | 0.399 | 0.4008 | 0.9006 | 0.8929 | 0.8966 | 0.4128 | 35.2636 |
| 1.0179 | 4.0 | 868 | 2.1396 | 0.479 | 0.2295 | 0.4121 | 0.4137 | 0.9024 | 0.8933 | 0.8977 | 0.4142 | 34.7182 |
| 0.8112 | 5.0 | 1085 | 2.2658 | 0.4737 | 0.2237 | 0.4046 | 0.405 | 0.8989 | 0.8931 | 0.8959 | 0.4126 | 35.4273 |
| 0.6745 | 6.0 | 1302 | 2.3615 | 0.4759 | 0.2278 | 0.407 | 0.4084 | 0.898 | 0.8937 | 0.8957 | 0.4311 | 37.0 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1