You can deploy this model and start earning money today!

cafierom/
bert-base-cased-finetuned-HMGCR-IC50s-V1

Name: bert-base-cased-finetuned-HMGCR-IC50s-V1
Availability: InStock

generated_from_trainer

google-bert/bert-base-cased

Views

No views yet

bert-base-cased-finetuned-HMGCR-IC50s-V1

This model is a fine-tuned version of bert-base-cased on 905 HMGCR IC50 values from bindingDB.org. Molecules with counter ions were included twice, once with and once without counter-ions.

It achieves the following results on the evaluation set:

Loss: 0.5334
Accuracy: 0.8214
F1: 0.8213

Use in Pipeline:

from transformers import pipeline

ic50_pipe = pipeline("text-classification", model="cafierom/bert-base-cased-finetuned-HMGCR-IC50s-V1")

bert_ic50 = ic50_pipe('COC(=O)C[C@H](O)C[C@H](O)\C=C\n1c(cc(c1-c1ccc(F)cc1

Model description

More information needed

Intended uses & limitations

Can classify HMGCR IC50 values as < 50 nM, < 500 nM, and > 500 nM. See Confusion matrix below:

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: 32
eval_batch_size: 32
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: linear
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.9409	1.0	25	0.8708	0.6929	0.6103
0.774	2.0	50	0.7855	0.6929	0.6153
0.6851	3.0	75	0.7056	0.7071	0.6251
0.576	4.0	100	0.7188	0.7143	0.6313
0.5388	5.0	125	0.6805	0.7429	0.7096
0.5123	6.0	150	0.6433	0.7571	0.7278
0.4618	7.0	175	0.6305	0.7357	0.6981
0.4575	8.0	200	0.6368	0.7429	0.7198
0.4304	9.0	225	0.6089	0.7643	0.7345
0.3834	10.0	250	0.6032	0.7714	0.7664
0.3928	11.0	275	0.6125	0.7714	0.7551
0.387	12.0	300	0.5994	0.8071	0.7930
0.3426	13.0	325	0.5814	0.7929	0.7798
0.3296	14.0	350	0.5230	0.8286	0.8266
0.298	15.0	375	0.5408	0.7786	0.7726
0.2997	16.0	400	0.5160	0.8	0.7980
0.2897	17.0	425	0.5344	0.8214	0.8213
0.2786	18.0	450	0.5534	0.7929	0.7872
0.262	19.0	475	0.5241	0.8	0.7957
0.2609	20.0	500	0.5256	0.8	0.7957

Framework versions

Transformers 4.48.3
Pytorch 2.5.1+cu124
Datasets 3.3.0
Tokenizers 0.21.0