nucleotide-transformer-v2-500m-multi-species_ft_BioS74_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of InstaDeepAI/nucleotide-transformer-v2-500m-multi-species on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.0901
• F1 Score: 0.8408
• Precision: 0.8543
• Recall: 0.8277
• Accuracy: 0.8359
• Auc: 0.9180
• Prc: 0.9190

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
0.5159	0.2629	500	0.4336	0.8179	0.7917	0.8458	0.8028	0.8805	0.8705
0.4166	0.5258	1000	0.4006	0.8434	0.7908	0.9036	0.8243	0.9079	0.9051
0.3854	0.7886	1500	0.3993	0.8481	0.7691	0.9453	0.8228	0.9127	0.9072
0.3755	1.0515	2000	0.3677	0.8348	0.8600	0.8112	0.8320	0.9252	0.9258
0.2996	1.3144	2500	0.3570	0.8506	0.8478	0.8533	0.8430	0.9268	0.9291
0.2827	1.5773	3000	0.3578	0.8564	0.8504	0.8624	0.8485	0.9305	0.9303
0.285	1.8402	3500	0.3630	0.8386	0.8674	0.8117	0.8364	0.9270	0.9280
0.2425	2.1030	4000	0.9870	0.8391	0.8578	0.8212	0.8351	0.9202	0.9203
0.4127	2.3659	4500	2.0901	0.8408	0.8543	0.8277	0.8359	0.9180	0.9190

Framework versions

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