prot_bert_classification_finetuned_karolina_es_10e

This model is a fine-tuned version of nepp1d0/prot_bert-finetuned-smiles-bindingDB on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6840
• Accuracy: 0.88
• F1: 0.9362
• Precision: 1.0
• Recall: 0.88

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-06
• train_batch_size: 32
• eval_batch_size: 32
• seed: 3
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	4	0.7082	0.02	0.0392	1.0	0.02
No log	2.0	8	0.7073	0.04	0.0769	1.0	0.04
No log	3.0	12	0.7060	0.04	0.0769	1.0	0.04
No log	4.0	16	0.7047	0.04	0.0769	1.0	0.04
No log	5.0	20	0.7034	0.08	0.1481	1.0	0.08
No log	6.0	24	0.7008	0.22	0.3607	1.0	0.22
No log	7.0	28	0.6976	0.22	0.3607	1.0	0.22
No log	8.0	32	0.6933	0.3	0.4615	1.0	0.3
No log	9.0	36	0.6893	0.6	0.7500	1.0	0.6
No log	10.0	40	0.6840	0.88	0.9362	1.0	0.88

Framework versions

• Transformers 4.23.1
• Pytorch 1.11.0
• Datasets 2.6.1
• Tokenizers 0.13.1