metadata
license: apache-2.0
base_model: google-t5/t5-small
tags:
- generated_from_trainer
datasets:
- scientific_papers
metrics:
- rouge
model-index:
- name: t5-small_arxiv_model
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: scientific_papers
type: scientific_papers
config: arxiv
split: test
args: arxiv
metrics:
- name: Rouge1
type: rouge
value: 0.1782
t5-small_arxiv_model
This model is a fine-tuned version of google-t5/t5-small on the scientific_papers dataset. It achieves the following results on the evaluation set:
- Loss: 2.5070
- Rouge1: 0.1782
- Rouge2: 0.0681
- Rougel: 0.1422
- Rougelsum: 0.1423
- Gen Len: 19.0
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.7744 | 1.0 | 20303 | 2.5639 | 0.1793 | 0.0691 | 0.1438 | 0.1439 | 19.0 |
| 2.6041 | 2.0 | 40606 | 2.5171 | 0.1778 | 0.0677 | 0.142 | 0.142 | 19.0 |
| 2.5843 | 3.0 | 60909 | 2.5070 | 0.1782 | 0.0681 | 0.1422 | 0.1423 | 19.0 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2