caduceus-ps_seqlen-131k_d_model-256_n_layer-16_ft_BioS74_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of kuleshov-group/caduceus-ps_seqlen-131k_d_model-256_n_layer-16 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8745
• F1 Score: 0.7577
• Precision: 0.7322
• Recall: 0.7850
• Accuracy: 0.7434
• Auc: 0.8008
• Prc: 0.8015

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Precision	Recall	Accuracy	Auc	Prc
0.5875	1.0	1902	0.5167	0.7580	0.7275	0.7912	0.7419	0.8081	0.7781
0.499	2.0	3804	0.4903	0.7743	0.7359	0.8169	0.7566	0.8188	0.7891
0.4781	3.0	5706	0.4782	0.7849	0.7402	0.8354	0.7660	0.8259	0.8059
0.4546	4.0	7608	0.4935	0.7839	0.7351	0.8395	0.7634	0.8222	0.7999
0.4228	5.0	9510	0.5101	0.7833	0.7422	0.8292	0.7655	0.8234	0.8134
0.3688	6.0	11412	0.5595	0.7704	0.7364	0.8076	0.7539	0.8129	0.8144
0.3289	7.0	13314	0.6433	0.7556	0.7319	0.7809	0.7419	0.8114	0.8089
0.2739	8.0	15216	0.8745	0.7577	0.7322	0.7850	0.7434	0.8008	0.8015

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.0
• Datasets 2.15.0
• Tokenizers 0.19.1