metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: Pegasus_paraphraser_Com_4
results: []
Pegasus_paraphraser_Com_4
This model is a fine-tuned version of Gayathri142214002/Pegasus_paraphraser_Com_3 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2817
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.2333 | 0.25 | 500 | 0.2292 |
| 0.2535 | 0.5 | 1000 | 0.2388 |
| 0.2607 | 0.75 | 1500 | 0.2452 |
| 0.2619 | 1.01 | 2000 | 0.2411 |
| 0.2038 | 1.26 | 2500 | 0.2596 |
| 0.2206 | 1.51 | 3000 | 0.2647 |
| 0.2228 | 1.76 | 3500 | 0.2642 |
| 0.2234 | 2.01 | 4000 | 0.2592 |
| 0.1747 | 2.26 | 4500 | 0.2771 |
| 0.1845 | 2.51 | 5000 | 0.2754 |
| 0.1968 | 2.77 | 5500 | 0.2672 |
| 0.1889 | 3.02 | 6000 | 0.2699 |
| 0.154 | 3.27 | 6500 | 0.2856 |
| 0.1596 | 3.52 | 7000 | 0.2827 |
| 0.1635 | 3.77 | 7500 | 0.2817 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3