nucleotide-transformer-2.5b-1000g_ft_BioS11_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of InstaDeepAI/nucleotide-transformer-2.5b-1000g on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8465
• F1 Score: 0.8645
• Precision: 0.8626
• Recall: 0.8663
• Accuracy: 0.8601
• Auc: 0.9345
• Prc: 0.9260

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: 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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Precision	Recall	Accuracy	Auc	Prc
0.4487	0.2753	500	0.3603	0.8644	0.7946	0.9476	0.8468	0.9261	0.9126
0.4104	0.5507	1000	0.4305	0.8635	0.7795	0.9679	0.8424	0.9284	0.9178
0.3731	0.8260	1500	0.3860	0.8672	0.7792	0.9775	0.8457	0.9362	0.9270
0.3218	1.1013	2000	0.4403	0.8745	0.8300	0.9241	0.8634	0.9357	0.9256
0.2396	1.3767	2500	0.5610	0.8717	0.8505	0.8941	0.8645	0.9359	0.9257
0.224	1.6520	3000	0.5519	0.8766	0.8473	0.9080	0.8683	0.9359	0.9266
0.2195	1.9273	3500	0.5169	0.8698	0.8647	0.8749	0.8650	0.9387	0.9319
0.0819	2.2026	4000	0.8884	0.8759	0.8351	0.9209	0.8656	0.9353	0.9257
0.0481	2.4780	4500	0.8794	0.8773	0.8160	0.9487	0.8634	0.9342	0.9248
0.0369	2.7533	5000	0.8465	0.8645	0.8626	0.8663	0.8601	0.9345	0.9260

Framework versions

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