metadata
license: apache-2.0
base_model: t5-small
tags:
- generated_from_trainer
datasets:
- samsum
metrics:
- rouge
model-index:
- name: t5-small-summarization-samsum
results: []
t5-small-summarization-samsum
This model is a fine-tuned version of t5-small on the samsum dataset. It achieves the following results on the evaluation set:
- Loss: 1.7907
- Rouge1: 0.4318
- Rouge2: 0.2005
- Rougel: 0.3629
- Rougelsum: 0.3629
- Gen Len: 16.8971
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: 16
- eval_batch_size: 16
- 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 | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.1074 | 0.54 | 500 | 1.9230 | 0.4011 | 0.182 | 0.3418 | 0.3417 | 15.7439 |
| 2.0526 | 1.09 | 1000 | 1.8559 | 0.4122 | 0.1841 | 0.3478 | 0.348 | 16.386 |
| 2.0075 | 1.63 | 1500 | 1.8193 | 0.4273 | 0.1955 | 0.3552 | 0.3551 | 16.8554 |
| 1.97 | 2.17 | 2000 | 1.8086 | 0.4222 | 0.1922 | 0.3551 | 0.3552 | 16.761 |
| 1.931 | 2.72 | 2500 | 1.7907 | 0.4318 | 0.2005 | 0.3629 | 0.3629 | 16.8971 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1