metadata
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
base_model: Rostlab/prot_bert
model-index:
- name: ProtBert-finetuned-proteinBindingDB
results: []
ProtBert-finetuned-proteinBindingDB
This model is a fine-tuned version of Rostlab/prot_bert on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5764
- Accuracy: 0.885
- F1: 0.8459
- Precision: 0.8255
- Recall: 0.885
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: 1e-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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.8056 | 1.0 | 5000 | 1.5153 | 0.745 | 0.6391 | 0.5606 | 0.745 |
| 0.7873 | 2.0 | 10000 | 0.5976 | 0.865 | 0.8267 | 0.8063 | 0.865 |
| 0.7427 | 3.0 | 15000 | 0.6316 | 0.875 | 0.8364 | 0.8176 | 0.875 |
| 1.0022 | 4.0 | 20000 | 0.6766 | 0.85 | 0.8112 | 0.7951 | 0.85 |
| 0.7379 | 5.0 | 25000 | 0.6181 | 0.865 | 0.8267 | 0.8063 | 0.865 |
| 0.6987 | 6.0 | 30000 | 0.7094 | 0.87 | 0.8336 | 0.82 | 0.87 |
| 0.6984 | 7.0 | 35000 | 0.5377 | 0.885 | 0.8471 | 0.8290 | 0.885 |
| 0.6657 | 8.0 | 40000 | 0.6278 | 0.875 | 0.8373 | 0.8213 | 0.875 |
| 0.6695 | 9.0 | 45000 | 0.6323 | 0.88 | 0.8421 | 0.8240 | 0.88 |
| 0.6352 | 10.0 | 50000 | 0.5764 | 0.885 | 0.8459 | 0.8255 | 0.885 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1