|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: t5-small-finetuned-NL2ModelioMQ |
|
results: [] |
|
--- |
|
|
|
<!-- 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-small-finetuned-NL2ModelioMQ |
|
|
|
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the ModelioMQ dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0000 |
|
- Rouge2 Precision: 0.9788 |
|
- Rouge2 Recall: 0.6053 |
|
- Rouge2 Fmeasure: 0.7294 |
|
|
|
## Model description |
|
|
|
|
|
## Intended uses & limitations |
|
|
|
This model will be used to assist [Modelio](https://www.modelio.org/) users in writing model queries in the ModelioMQ language. |
|
|
|
## Training and evaluation data |
|
|
|
The ModelioMQ dataset originates from 79 manually-written NL|ModelioMQ pairs, which have been augmentated using [NLPAug](https://github.com/makcedward/nlpaug) |
|
|
|
## 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 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure | |
|
|:-------------:|:-----:|:-----:|:---------------:|:----------------:|:-------------:|:---------------:| |
|
| 0.0104 | 1.0 | 4449 | 0.0006 | 0.9699 | 0.601 | 0.7235 | |
|
| 0.002 | 2.0 | 8898 | 0.0000 | 0.9788 | 0.6053 | 0.7294 | |
|
| 0.001 | 3.0 | 13347 | 0.0000 | 0.9788 | 0.6053 | 0.7294 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.24.0 |
|
- Pytorch 1.12.1+cu113 |
|
- Tokenizers 0.13.2 |
|
|
