bert-base-cased-ChemTok-ZN15-55KTyrosinase-V1-finetuned-Tyrosinase-IC50s-V3

This model is a fine-tuned version of cafierom/bert-base-cased-ChemTok-ZN15-55KTyrosinase-V1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7157
• Accuracy: 0.7473
• F1: 0.7495

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.0862	1.0	17	1.0233	0.5108	0.5118
1.003	2.0	34	0.8975	0.6237	0.6222
0.92	3.0	51	0.8070	0.6398	0.6369
0.8326	4.0	68	0.7641	0.6505	0.6505
0.7697	5.0	85	0.7158	0.6935	0.6909
0.6863	6.0	102	0.7421	0.6667	0.6578
0.6336	7.0	119	0.7156	0.6935	0.6925
0.5629	8.0	136	0.7755	0.6828	0.6819
0.528	9.0	153	0.7701	0.6667	0.6707
0.4968	10.0	170	0.7473	0.7043	0.7019
0.4814	11.0	187	0.7157	0.7473	0.7495
0.4264	12.0	204	0.7831	0.6935	0.6882
0.4154	13.0	221	0.8320	0.6667	0.6598
0.3749	14.0	238	0.8113	0.6774	0.6781
0.3287	15.0	255	0.8504	0.7097	0.7122
0.3533	16.0	272	0.8572	0.7204	0.7204
0.323	17.0	289	1.0137	0.6774	0.6771
0.3017	18.0	306	0.8927	0.7151	0.7132
0.2777	19.0	323	0.8790	0.7151	0.7161
0.2752	20.0	340	0.9587	0.6828	0.6838

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1