license: apache-2.0 | |
tags: | |
- generated_from_trainer | |
datasets: | |
- super_glue | |
metrics: | |
- rouge | |
model-index: | |
- name: T5_finetuned | |
results: | |
- task: | |
name: Sequence-to-sequence Language Modeling | |
type: text2text-generation | |
dataset: | |
name: super_glue | |
type: super_glue | |
config: boolq | |
split: train | |
args: boolq | |
metrics: | |
- name: Rouge1 | |
type: rouge | |
value: 79.3272 | |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You | |
should probably proofread and complete it, then remove this comment. --> | |
# T5_finetuned | |
This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the super_glue dataset. | |
It achieves the following results on the evaluation set: | |
- Loss: 0.1077 | |
- Rouge1: 79.3272 | |
- Rouge2: 0.0 | |
- Rougel: 79.2966 | |
- Rougelsum: 79.3272 | |
- Gen Len: 2.8269 | |
## 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: 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 | | |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:| | |
| 0.5134 | 1.0 | 590 | 0.1102 | 79.8165 | 0.0 | 79.8165 | 79.8471 | 2.7713 | | |
| 0.105 | 2.0 | 1180 | 0.1049 | 80.3364 | 0.0 | 80.3364 | 80.367 | 2.6483 | | |
| 0.1023 | 3.0 | 1770 | 0.1077 | 79.3272 | 0.0 | 79.2966 | 79.3272 | 2.8269 | | |
### Framework versions | |
- Transformers 4.25.1 | |
- Pytorch 1.13.0+cu116 | |
- Datasets 2.8.0 | |
- Tokenizers 0.13.2 | |