metadata
library_name: transformers
license: apache-2.0
base_model: google-t5/t5-small
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: podcast_summarizer
results: []
podcast_summarizer
This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.1084
- Rouge1: 0.1735
- Rouge2: 0.0673
- Rougel: 0.1411
- Rougelsum: 0.1413
- Gen Len: 18.9721
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 179 | 2.2171 | 0.1569 | 0.0555 | 0.1271 | 0.1276 | 18.9721 |
| No log | 2.0 | 358 | 2.1460 | 0.1653 | 0.0612 | 0.1346 | 0.1348 | 18.9721 |
| 2.3662 | 3.0 | 537 | 2.1163 | 0.1723 | 0.066 | 0.1398 | 0.1398 | 18.9721 |
| 2.3662 | 4.0 | 716 | 2.1084 | 0.1735 | 0.0673 | 0.1411 | 0.1413 | 18.9721 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu121
- Tokenizers 0.20.3