|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: med_masked_pubmed_articles_biogpt |
|
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. --> |
|
|
|
# med_masked_pubmed_articles_biogpt |
|
|
|
This model is a fine-tuned version of [microsoft/biogpt](https://huggingface.co/microsoft/biogpt) on a med_masked_pubmed_articles dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.1952 |
|
- Rouge2 Precision: 0.7072 |
|
- Rouge2 Recall: 0.7001 |
|
- Rouge2 Fmeasure: 0.7025 |
|
|
|
## 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: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure | |
|
|:-------------:|:-----:|:-----:|:---------------:|:----------------:|:-------------:|:---------------:| |
|
| 3.1392 | 1.0 | 7914 | 3.0945 | 0.7075 | 0.7001 | 0.7026 | |
|
| 2.927 | 2.0 | 15828 | 3.0705 | 0.7074 | 0.7001 | 0.7026 | |
|
| 2.8558 | 3.0 | 23742 | 3.0877 | 0.7073 | 0.7001 | 0.7025 | |
|
| 2.7035 | 4.0 | 31656 | 3.1354 | 0.7073 | 0.7001 | 0.7026 | |
|
| 2.6209 | 5.0 | 39570 | 3.1952 | 0.7072 | 0.7001 | 0.7025 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.13.2 |
|
|
