caduceus-ps_seqlen-131k_d_model-256_n_layer-16_ft_BioS2_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: 405698528.0
• F1 Score: 0.0
• Precision: 0.0
• Recall: 0.0
• Accuracy: 0.4691
• Auc: 0.3381
• Prc: 0.4187

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
40362173267.968	1.0	2974	19486085120.0	0.7290	0.6012	0.9259	0.6345	0.3721	0.4272
8689587060.736	2.0	5948	7442824704.0	0.3919	0.5643	0.3002	0.5054	0.3808	0.4313
4078919745.536	3.0	8922	2803392256.0	0.6194	0.5420	0.7226	0.5286	0.3380	0.4145
825952436.224	4.0	11896	1180509824.0	0.0013	0.2	0.0006	0.4681	0.2804	0.3953
507201716.224	5.0	14870	550735680.0	0.6938	0.5311	1.0	0.5313	0.2979	0.4028
277415460.864	6.0	17844	81736368.0	0.7489	0.6446	0.8936	0.6819	0.3057	0.4062
283333623.808	7.0	20818	350532352.0	0.6936	0.5309	1.0	0.5309	0.3136	0.4098
615652327.424	8.0	23792	84891424.0	0.7110	0.5615	0.9690	0.5817	0.3362	0.4203
310394716.16	9.0	26766	629298560.0	0.0	0.0	0.0	0.4684	0.3425	0.4226
438004056.064	10.0	29740	143799264.0	0.6938	0.5311	1.0	0.5313	0.3390	0.4212
201700261.888	11.0	32714	405698528.0	0.0	0.0	0.0	0.4691	0.3381	0.4187

Framework versions

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