---
license: bsd-3-clause
base_model: LongSafari/hyenadna-small-32k-seqlen-hf
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: hyenadna-small-32k-seqlen-hf_ft_BioS45_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-small-32k-seqlen-hf_ft_BioS45_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of [LongSafari/hyenadna-small-32k-seqlen-hf](https://huggingface.co/LongSafari/hyenadna-small-32k-seqlen-hf) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5632
- F1 Score: 0.8023
- Precision: 0.8234
- Recall: 0.7823
- Accuracy: 0.7989
- Auc: 0.8729
- Prc: 0.8701

## 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    |
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|:------:|:------:|
| 0.5477        | 0.2103 | 500  | 0.4845          | 0.7868   | 0.7923    | 0.7815 | 0.7791   | 0.8482 | 0.8295 |
| 0.4789        | 0.4207 | 1000 | 0.4609          | 0.8003   | 0.7943    | 0.8065 | 0.7901   | 0.8624 | 0.8547 |
| 0.4505        | 0.6310 | 1500 | 0.4569          | 0.7972   | 0.8068    | 0.7879 | 0.7909   | 0.8700 | 0.8670 |
| 0.4558        | 0.8414 | 2000 | 0.4471          | 0.8289   | 0.7664    | 0.9024 | 0.8056   | 0.8753 | 0.8637 |
| 0.4376        | 1.0517 | 2500 | 0.4435          | 0.8324   | 0.7840    | 0.8871 | 0.8136   | 0.8786 | 0.8708 |
| 0.4299        | 1.2621 | 3000 | 0.4596          | 0.8284   | 0.7795    | 0.8839 | 0.8090   | 0.8770 | 0.8676 |
| 0.4329        | 1.4724 | 3500 | 0.4409          | 0.8133   | 0.8203    | 0.8065 | 0.8069   | 0.8805 | 0.8781 |
| 0.4187        | 1.6828 | 4000 | 0.4495          | 0.8063   | 0.8246    | 0.7887 | 0.8023   | 0.8802 | 0.8798 |
| 0.4211        | 1.8931 | 4500 | 0.4363          | 0.8303   | 0.7925    | 0.8718 | 0.8141   | 0.8807 | 0.8734 |
| 0.3818        | 2.1035 | 5000 | 0.4506          | 0.8350   | 0.7932    | 0.8815 | 0.8183   | 0.8809 | 0.8717 |
| 0.3524        | 2.3138 | 5500 | 0.4775          | 0.8184   | 0.7968    | 0.8411 | 0.8052   | 0.8785 | 0.8745 |
| 0.3601        | 2.5242 | 6000 | 0.4678          | 0.8263   | 0.7770    | 0.8823 | 0.8065   | 0.8758 | 0.8688 |
| 0.3708        | 2.7345 | 6500 | 0.4836          | 0.8226   | 0.8064    | 0.8395 | 0.8111   | 0.8769 | 0.8743 |
| 0.3911        | 2.9449 | 7000 | 0.4733          | 0.8207   | 0.7912    | 0.8524 | 0.8056   | 0.8748 | 0.8666 |
| 0.3136        | 3.1552 | 7500 | 0.5632          | 0.8023   | 0.8234    | 0.7823 | 0.7989   | 0.8729 | 0.8701 |


### Framework versions

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