nucleotide-transformer-2.5b-multi-species_ft_BioS2_1kbpHG19_DHSs_H3K27AC

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

• Loss: 0.3464
• F1 Score: 0.8617
• Precision: 0.8829
• Recall: 0.8414
• Accuracy: 0.8570
• Auc: 0.9374
• Prc: 0.9355

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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Precision	Recall	Accuracy	Auc	Prc
0.5131	0.0841	500	0.4263	0.8248	0.7933	0.8589	0.8069	0.8829	0.8773
0.453	0.1682	1000	0.4288	0.8085	0.8500	0.7708	0.8068	0.8988	0.8974
0.4295	0.2522	1500	0.4352	0.8478	0.7642	0.9520	0.8192	0.9110	0.9080
0.4114	0.3363	2000	0.3765	0.8308	0.8743	0.7915	0.8295	0.9198	0.9167
0.3881	0.4204	2500	0.4381	0.8181	0.8932	0.7546	0.8224	0.9225	0.9235
0.3833	0.5045	3000	0.3577	0.8631	0.8023	0.9339	0.8433	0.9255	0.9247
0.3849	0.5885	3500	0.3715	0.8548	0.8392	0.8709	0.8434	0.9223	0.9209
0.3506	0.6726	4000	0.4040	0.8657	0.8452	0.8872	0.8544	0.9288	0.9261
0.3708	0.7567	4500	0.3428	0.8619	0.8731	0.8509	0.8557	0.9329	0.9309
0.3581	0.8408	5000	0.4926	0.8216	0.9180	0.7435	0.8291	0.9347	0.9331
0.3629	0.9248	5500	0.3464	0.8617	0.8829	0.8414	0.8570	0.9374	0.9355

Framework versions

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