metadata

license: cc-by-nc-sa-4.0
base_model: InstaDeepAI/nucleotide-transformer-v2-100m-multi-species
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: >-
      nucleotide-transformer-v2-100m-multi-species_ft_BioS74_1kbpHG19_DHSs_H3K27AC
    results: []

nucleotide-transformer-v2-100m-multi-species_ft_BioS74_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of InstaDeepAI/nucleotide-transformer-v2-100m-multi-species on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.4219
F1 Score: 0.8409
Precision: 0.8361
Recall: 0.8458
Accuracy: 0.8325
Auc: 0.9111
Prc: 0.9049

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
0.5544	0.1314	500	0.5244	0.7566	0.7764	0.7378	0.7515	0.8400	0.8313
0.5015	0.2629	1000	0.4997	0.7709	0.824	0.7243	0.7747	0.8691	0.8632
0.4614	0.3943	1500	0.4311	0.8219	0.8011	0.8438	0.8086	0.8891	0.8855
0.4294	0.5258	2000	0.4329	0.8290	0.7769	0.8885	0.8080	0.8935	0.8874
0.4045	0.6572	2500	0.4249	0.8331	0.7673	0.9111	0.8088	0.8987	0.8937
0.432	0.7886	3000	0.4033	0.8261	0.8244	0.8277	0.8175	0.8993	0.8931
0.4148	0.9201	3500	0.4032	0.8381	0.8156	0.8619	0.8257	0.9057	0.8986
0.4127	1.0515	4000	0.4252	0.8196	0.8417	0.7986	0.8159	0.9026	0.8932
0.3807	1.1830	4500	0.3981	0.8441	0.8146	0.8759	0.8307	0.9046	0.8971
0.3711	1.3144	5000	0.4066	0.8430	0.8130	0.8754	0.8293	0.9011	0.8941
0.363	1.4458	5500	0.5295	0.8012	0.8737	0.7398	0.8078	0.9053	0.9037
0.3624	1.5773	6000	0.4219	0.8409	0.8361	0.8458	0.8325	0.9111	0.9049

Framework versions

Transformers 4.42.3
Pytorch 2.3.0+cu121
Datasets 2.18.0
Tokenizers 0.19.0