metadata
library_name: transformers
license: apache-2.0
base_model: sshleifer/distilbart-xsum-12-6
tags:
- generated_from_trainer
model-index:
- name: bart-abs-2409-1947-lr-0.0003-bs-8-maxep-6
results: []
bart-abs-2409-1947-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: 7.3898
- Rouge/rouge1: 0.3035
- Rouge/rouge2: 0.072
- Rouge/rougel: 0.2428
- Rouge/rougelsum: 0.2429
- Bertscore/bertscore-precision: 0.8724
- Bertscore/bertscore-recall: 0.8571
- Bertscore/bertscore-f1: 0.8646
- Meteor: 0.2108
- Gen Len: 29.0
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.3901 | 1.0 | 109 | 6.5833 | 0.2377 | 0.0309 | 0.193 | 0.1932 | 0.8496 | 0.853 | 0.8512 | 0.2159 | 45.0 |
| 0.3274 | 2.0 | 218 | 6.5583 | 0.2439 | 0.0504 | 0.2065 | 0.2067 | 0.8544 | 0.8581 | 0.8562 | 0.229 | 45.0 |
| 0.3098 | 3.0 | 327 | 6.9294 | 0.2613 | 0.0803 | 0.214 | 0.2142 | 0.8711 | 0.8469 | 0.8588 | 0.2102 | 25.0 |
| 0.2625 | 4.0 | 436 | 7.0223 | 0.3008 | 0.0767 | 0.229 | 0.2292 | 0.858 | 0.8674 | 0.8626 | 0.2167 | 41.0 |
| 0.2379 | 5.0 | 545 | 7.2276 | 0.3035 | 0.072 | 0.2428 | 0.2429 | 0.8724 | 0.8571 | 0.8646 | 0.2108 | 29.0 |
| 0.2168 | 6.0 | 654 | 7.3898 | 0.3035 | 0.072 | 0.2428 | 0.2429 | 0.8724 | 0.8571 | 0.8646 | 0.2108 | 29.0 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0
- Datasets 3.0.0
- Tokenizers 0.19.1