README.md

metadata

license: cc-by-nc-sa-4.0
base_model: InstaDeepAI/nucleotide-transformer-500m-human-ref
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: nucleotide-transformer-500m-human-ref_ft_BioS2_1kbpHG19_DHSs_H3K27AC
    results: []

nucleotide-transformer-500m-human-ref_ft_BioS2_1kbpHG19_DHSs_H3K27AC

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

• Loss: 0.4074
• F1 Score: 0.8357
• Precision: 0.8177
• Recall: 0.8545
• Accuracy: 0.8268
• Auc: 0.8999
• Prc: 0.8854

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
0.5531	0.0841	500	0.4955	0.7552	0.8200	0.6999	0.7661	0.8646	0.8453
0.4826	0.1681	1000	0.4657	0.7900	0.8043	0.7763	0.7873	0.8686	0.8463
0.469	0.2522	1500	0.4324	0.8114	0.8090	0.8138	0.8050	0.8829	0.8696
0.4508	0.3362	2000	0.4388	0.8072	0.8311	0.7847	0.8068	0.8906	0.8743
0.4562	0.4203	2500	0.4340	0.8304	0.7769	0.8917	0.8122	0.8902	0.8798
0.4393	0.5044	3000	0.4291	0.8330	0.7695	0.9080	0.8124	0.8946	0.8823
0.4368	0.5884	3500	0.4122	0.8382	0.7903	0.8924	0.8225	0.8971	0.8839
0.4354	0.6725	4000	0.4191	0.8359	0.7751	0.9070	0.8164	0.8953	0.8821
0.4346	0.7566	4500	0.4110	0.8326	0.8113	0.8552	0.8228	0.8991	0.8874
0.4298	0.8406	5000	0.4074	0.8357	0.8177	0.8545	0.8268	0.8999	0.8854

Framework versions

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