|
|
--- |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- knkarthick/samsum |
|
|
language: |
|
|
- en |
|
|
metrics: |
|
|
- rouge |
|
|
base_model: |
|
|
- google-t5/t5-base |
|
|
pipeline_tag: summarization |
|
|
tags: |
|
|
- generated_from_text |
|
|
--- |
|
|
## Summarizer |
|
|
This model is a fine-tuned version of google-t5/t5-base on a knkarthick/samsum dataset. It achieves the following results on the evaluation set: |
|
|
|
|
|
• ROUGE-1: 51.41 |
|
|
|
|
|
• ROUGE-2: 26.72 |
|
|
|
|
|
• ROUGE-L: 42.15 |
|
|
|
|
|
• ROUGE-Lsum: 42.17 |
|
|
|
|
|
## Training hyperparameters |
|
|
|
|
|
The following hyperparameters were used during training: |
|
|
|
|
|
• learning_rate: 5e-5 |
|
|
|
|
|
• train_batch_size: 2 |
|
|
|
|
|
• gradient_accumulation_steps=2 |
|
|
|
|
|
• seed: 42 |
|
|
|
|
|
• weight_decay=0.01 |
|
|
|
|
|
• lr_scheduler_type: linear |
|
|
|
|
|
• warmup_ratio=0.1 |
|
|
|
|
|
• num_epochs: 3 |
|
|
|
|
|
## Training results |
|
|
| Training Loss | Epoch | Step | |
|
|
| ------------- | ----- | ----- | |
|
|
| 1.491800 | 1 | 3650 | |
|
|
| 1.404900 | 2 | 7350 | |
|
|
| 1.322800 | 3 | 11000 | |
|
|
|
|
|
|
|
|
## Framework versions |
|
|
• transformers: 4.56.0 |
|
|
|
|
|
• pytorch: 2.0.1+cu118 |
|
|
|
|
|
• datasets: 2.14.4 |
|
|
|
|
|
• tokenizers: 0.13.3 |
|
|
|
