---
license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
- audio-classification
- hubert
- esc50
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: hubert-esc50-finetuned-v2
  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. -->

# hubert-esc50-finetuned-v2

This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on the ESC-50 dataset.
It achieves the following results on the evaluation set:
- Loss: 1.8681
- Accuracy: 0.5175

## 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: 5e-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: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 3.5459        | 1.0   | 200  | 3.5354          | 0.09     |
| 3.1002        | 2.0   | 400  | 3.2000          | 0.1375   |
| 2.8786        | 3.0   | 600  | 2.8955          | 0.175    |
| 2.5762        | 4.0   | 800  | 2.6800          | 0.2325   |
| 2.3834        | 5.0   | 1000 | 2.5289          | 0.2725   |
| 2.3231        | 6.0   | 1200 | 2.2412          | 0.375    |
| 1.9827        | 7.0   | 1400 | 2.0902          | 0.435    |
| 1.9386        | 8.0   | 1600 | 1.9596          | 0.4725   |
| 1.7323        | 9.0   | 1800 | 1.9183          | 0.4675   |
| 1.7958        | 10.0  | 2000 | 1.8681          | 0.5175   |


### Framework versions

- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1