---
license: bsd-3-clause
base_model: LongSafari/hyenadna-medium-450k-seqlen-hf
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: hyenadna-medium-450k-seqlen-hf_ft_BioS73_1kbpHG19_DHSs_H3K27AC
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# hyenadna-medium-450k-seqlen-hf_ft_BioS73_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of [LongSafari/hyenadna-medium-450k-seqlen-hf](https://huggingface.co/LongSafari/hyenadna-medium-450k-seqlen-hf) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4797
- F1 Score: 0.8324
- Precision: 0.8063
- Recall: 0.8603
- Accuracy: 0.8151
- Auc: 0.8805
- Prc: 0.8644

## 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: 16
- eval_batch_size: 16
- 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    |
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|:------:|:------:|
| 0.4819        | 0.3726 | 500  | 0.4409          | 0.8212   | 0.7769    | 0.8708 | 0.7976   | 0.8718 | 0.8657 |
| 0.4306        | 0.7452 | 1000 | 0.4281          | 0.8181   | 0.8103    | 0.8261 | 0.8040   | 0.8826 | 0.8790 |
| 0.4073        | 1.1177 | 1500 | 0.4115          | 0.8310   | 0.7977    | 0.8673 | 0.8118   | 0.8884 | 0.8864 |
| 0.3903        | 1.4903 | 2000 | 0.4011          | 0.8345   | 0.8096    | 0.8610 | 0.8177   | 0.8902 | 0.8832 |
| 0.3808        | 1.8629 | 2500 | 0.4221          | 0.8295   | 0.8075    | 0.8527 | 0.8129   | 0.8865 | 0.8788 |
| 0.3713        | 2.2355 | 3000 | 0.4126          | 0.8305   | 0.8212    | 0.8401 | 0.8170   | 0.8959 | 0.8883 |
| 0.3587        | 2.6080 | 3500 | 0.4023          | 0.8469   | 0.7937    | 0.9078 | 0.8248   | 0.8936 | 0.8857 |
| 0.3362        | 2.9806 | 4000 | 0.4117          | 0.8489   | 0.7852    | 0.9239 | 0.8245   | 0.8925 | 0.8829 |
| 0.3075        | 3.3532 | 4500 | 0.4235          | 0.8295   | 0.8120    | 0.8478 | 0.8140   | 0.8946 | 0.8877 |
| 0.319         | 3.7258 | 5000 | 0.4154          | 0.8497   | 0.8096    | 0.8939 | 0.8312   | 0.8959 | 0.8868 |
| 0.2986        | 4.0984 | 5500 | 0.4481          | 0.8377   | 0.8028    | 0.8757 | 0.8189   | 0.8929 | 0.8835 |
| 0.2712        | 4.4709 | 6000 | 0.4398          | 0.8281   | 0.8005    | 0.8575 | 0.8099   | 0.8833 | 0.8708 |
| 0.2819        | 4.8435 | 6500 | 0.4465          | 0.8324   | 0.8189    | 0.8464 | 0.8181   | 0.8904 | 0.8828 |
| 0.245         | 5.2161 | 7000 | 0.4827          | 0.8294   | 0.8125    | 0.8471 | 0.8140   | 0.8853 | 0.8713 |
| 0.2318        | 5.5887 | 7500 | 0.4797          | 0.8324   | 0.8063    | 0.8603 | 0.8151   | 0.8805 | 0.8644 |


### Framework versions

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