metadata

base_model: Rostlab/prot_bert
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: prot_bert-fine-tuned-toxicity_3
    results: []

prot_bert-fine-tuned-toxicity_3

This model is a fine-tuned version of Rostlab/prot_bert on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.0450
Accuracy: 0.7419
Precision: 0.7530
Recall: 0.7419
F1: 0.7340

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: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.5735	1.0	16	0.5968	0.8065	0.8093	0.8065	0.8044
0.4855	2.0	32	0.4939	0.8387	0.8391	0.8387	0.8380
0.3499	3.0	48	0.5234	0.8065	0.8093	0.8065	0.8044
0.3094	4.0	64	0.4639	0.8387	0.8391	0.8387	0.8380
0.235	5.0	80	0.5654	0.8387	0.8391	0.8387	0.8380
0.1608	6.0	96	0.6409	0.8387	0.8391	0.8387	0.8380
0.1147	7.0	112	0.6271	0.8387	0.8391	0.8387	0.8380
0.171	8.0	128	0.7704	0.8065	0.8093	0.8065	0.8044
0.1762	9.0	144	0.7978	0.8065	0.8093	0.8065	0.8044
0.1438	10.0	160	0.9561	0.7419	0.7530	0.7419	0.7340
0.0953	11.0	176	1.0074	0.7419	0.7530	0.7419	0.7340
0.0394	12.0	192	1.0037	0.7419	0.7530	0.7419	0.7340
0.0896	13.0	208	1.0128	0.7419	0.7530	0.7419	0.7340
0.0849	14.0	224	1.0404	0.7419	0.7530	0.7419	0.7340
0.0207	15.0	240	1.0450	0.7419	0.7530	0.7419	0.7340

Framework versions

Transformers 4.41.2
Pytorch 2.3.0+cu121
Datasets 2.20.0
Tokenizers 0.19.1