|
--- |
|
base_model: Rostlab/prot_bert |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
- f1 |
|
model-index: |
|
- name: prot_bert-fine-tuned-toxicity_2.0.1 |
|
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. --> |
|
|
|
# prot_bert-fine-tuned-toxicity_2.0.1 |
|
|
|
This model is a fine-tuned version of [Rostlab/prot_bert](https://huggingface.co/Rostlab/prot_bert) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4239 |
|
- Accuracy: 0.8387 |
|
- Precision: 0.8391 |
|
- Recall: 0.8387 |
|
- F1: 0.8380 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 8 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:| |
|
| 0.6945 | 1.0 | 16 | 0.6730 | 0.7742 | 0.8401 | 0.7742 | 0.7558 | |
|
| 0.6185 | 2.0 | 32 | 0.6087 | 0.8065 | 0.8093 | 0.8065 | 0.8044 | |
|
| 0.5241 | 3.0 | 48 | 0.5501 | 0.8387 | 0.8391 | 0.8387 | 0.8380 | |
|
| 0.466 | 4.0 | 64 | 0.5328 | 0.8387 | 0.8391 | 0.8387 | 0.8380 | |
|
| 0.4281 | 5.0 | 80 | 0.6886 | 0.7097 | 0.7252 | 0.7097 | 0.6967 | |
|
| 0.3393 | 6.0 | 96 | 0.4568 | 0.8387 | 0.8391 | 0.8387 | 0.8380 | |
|
| 0.1909 | 7.0 | 112 | 0.4239 | 0.8387 | 0.8391 | 0.8387 | 0.8380 | |
|
| 0.243 | 8.0 | 128 | 0.4323 | 0.8387 | 0.8391 | 0.8387 | 0.8380 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
|
|
