metadata
license: apache-2.0
base_model: facebook/bart-large
tags:
- generated_from_trainer
datasets:
- arrow
model-index:
- name: Longformer_BART_dependent_V1
results: []
Longformer_BART_dependent_V1
This model is a fine-tuned version of facebook/bart-large on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 2.0684
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 40
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.9474 | 1.0 | 69 | 2.3410 |
| 2.4046 | 2.0 | 138 | 2.1455 |
| 2.1646 | 3.0 | 207 | 2.0931 |
| 2.0241 | 4.0 | 276 | 2.0488 |
| 1.8985 | 5.0 | 345 | 2.0295 |
| 1.7871 | 6.0 | 414 | 2.0230 |
| 1.7065 | 7.0 | 483 | 2.0143 |
| 1.6223 | 8.0 | 552 | 2.0259 |
| 1.5524 | 9.0 | 621 | 2.0266 |
| 1.4814 | 10.0 | 690 | 2.0458 |
| 1.4187 | 11.0 | 759 | 2.0515 |
| 1.3658 | 12.0 | 828 | 2.0684 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.19.1