README.md

metadata

base_model: zhihan1996/DNABERT-2-117M
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: DNABERT-2-117M_ft_BioS2_1kbpHG19_DHSs_H3K27AC
    results: []

DNABERT-2-117M_ft_BioS2_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of zhihan1996/DNABERT-2-117M on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4620
• F1 Score: 0.7974
• Precision: 0.8134
• Recall: 0.7819
• Accuracy: 0.7933
• Auc: 0.8741
• Prc: 0.8601

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Precision	Recall	Accuracy	Auc	Prc
0.706	0.1680	500	0.6216	0.6107	0.7600	0.5105	0.6616	0.7690	0.7525
0.614	0.3360	1000	0.5784	0.7600	0.6685	0.8805	0.7108	0.7836	0.7638
0.5795	0.5040	1500	0.5409	0.7576	0.6964	0.8307	0.7236	0.7938	0.7723
0.557	0.6720	2000	0.5898	0.7622	0.6293	0.9661	0.6864	0.8063	0.7838
0.561	0.8401	2500	0.5143	0.7770	0.7298	0.8307	0.7520	0.8183	0.7953
0.5313	1.0081	3000	0.4928	0.7958	0.7300	0.8746	0.7666	0.8334	0.8096
0.5167	1.1761	3500	0.5086	0.7590	0.7963	0.7250	0.7605	0.8471	0.8266
0.5052	1.3441	4000	0.4732	0.8021	0.7578	0.8520	0.7814	0.8533	0.8327
0.4923	1.5121	4500	0.4741	0.7898	0.7864	0.7932	0.7804	0.8572	0.8386
0.493	1.6801	5000	0.4683	0.7821	0.7995	0.7654	0.7782	0.8618	0.8431
0.4834	1.8481	5500	0.4522	0.8100	0.7702	0.8543	0.7916	0.8638	0.8465
0.4871	2.0161	6000	0.4527	0.8160	0.7475	0.8982	0.7893	0.8665	0.8500
0.457	2.1841	6500	0.6625	0.7653	0.6231	0.9916	0.6838	0.8679	0.8530
0.4685	2.3522	7000	0.5230	0.7281	0.8393	0.6430	0.7503	0.8681	0.8521
0.4567	2.5202	7500	0.4433	0.8198	0.7557	0.8956	0.7952	0.8702	0.8533
0.4506	2.6882	8000	0.4520	0.8208	0.7520	0.9034	0.7948	0.8721	0.8580
0.4443	2.8562	8500	0.4590	0.8029	0.7963	0.8097	0.7933	0.8730	0.8581
0.4433	3.0242	9000	0.5848	0.7530	0.8380	0.6837	0.7668	0.8738	0.8592
0.4355	3.1922	9500	0.4620	0.7974	0.8134	0.7819	0.7933	0.8741	0.8601

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.0+cu121
• Datasets 2.18.0
• Tokenizers 0.19.0