Geneformer_ft_BioS2_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of ctheodoris/Geneformer on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6514
• F1 Score: 0.6398
• Precision: 0.6615
• Recall: 0.6195
• Accuracy: 0.6273
• Auc: 0.6654
• Prc: 0.6684

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
0.6959	0.0841	500	0.6846	0.6785	0.5633	0.8530	0.5681	0.5765	0.5859
0.6879	0.1682	1000	0.6892	0.3661	0.6392	0.2565	0.5254	0.5823	0.6044
0.6913	0.2523	1500	0.6919	0.2450	0.6413	0.1514	0.5013	0.5921	0.6000
0.6854	0.3364	2000	0.6722	0.6928	0.5659	0.8930	0.5769	0.6111	0.6274
0.6774	0.4205	2500	0.6637	0.6896	0.5917	0.8263	0.6026	0.6321	0.6402
0.6667	0.5045	3000	0.6546	0.6895	0.6083	0.7957	0.6171	0.6509	0.6561
0.6791	0.5886	3500	0.6514	0.6398	0.6615	0.6195	0.6273	0.6654	0.6684

Framework versions

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