|
--- |
|
license: apache-2.0 |
|
base_model: facebook/bart-large |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: finetuned-bartL-samsum |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# finetuned-bartL-samsum |
|
|
|
This model is a fine-tuned version of [facebook/bart-large](https://huggingface.co/facebook/bart-large) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3301 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 2.0651 | 0.2 | 1000 | 0.5671 | |
|
| 0.4529 | 0.4 | 2000 | 0.4081 | |
|
| 0.4316 | 0.6 | 3000 | 0.3714 | |
|
| 0.4115 | 0.8 | 4000 | 0.3925 | |
|
| 0.3922 | 1.0 | 5000 | 0.3621 | |
|
| 0.3011 | 1.2 | 6000 | 0.3613 | |
|
| 0.3129 | 1.4 | 7000 | 0.3482 | |
|
| 0.2939 | 1.6 | 8000 | 0.3582 | |
|
| 0.2931 | 1.8 | 9000 | 0.3388 | |
|
| 0.2866 | 2.0 | 10000 | 0.3342 | |
|
| 0.2095 | 2.2 | 11000 | 0.3379 | |
|
| 0.2095 | 2.4 | 12000 | 0.3353 | |
|
| 0.2068 | 2.6 | 13000 | 0.3335 | |
|
| 0.2043 | 2.8 | 14000 | 0.3310 | |
|
| 0.1961 | 3.0 | 15000 | 0.3301 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
