|
--- |
|
base_model: haesun/pegasus-samsum |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: pegasus-ft |
|
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. --> |
|
|
|
# pegasus-ft |
|
|
|
This model is a fine-tuned version of [haesun/pegasus-samsum](https://huggingface.co/haesun/pegasus-samsum) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.9779 |
|
|
|
## 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: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.1074 | 0.19 | 100 | 2.0692 | |
|
| 1.9296 | 0.37 | 200 | 2.0481 | |
|
| 1.9456 | 0.56 | 300 | 2.0364 | |
|
| 1.9674 | 0.74 | 400 | 2.0205 | |
|
| 2.0887 | 0.93 | 500 | 2.0043 | |
|
| 2.0733 | 1.11 | 600 | 2.0023 | |
|
| 1.9111 | 1.3 | 700 | 1.9991 | |
|
| 1.887 | 1.48 | 800 | 1.9917 | |
|
| 2.0347 | 1.67 | 900 | 1.9855 | |
|
| 1.8488 | 1.85 | 1000 | 1.9838 | |
|
| 1.7699 | 2.04 | 1100 | 1.9800 | |
|
| 1.7386 | 2.22 | 1200 | 1.9795 | |
|
| 1.8828 | 2.41 | 1300 | 1.9804 | |
|
| 1.8072 | 2.59 | 1400 | 1.9808 | |
|
| 1.898 | 2.78 | 1500 | 1.9785 | |
|
| 1.9452 | 2.96 | 1600 | 1.9779 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|