---
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan
  results:
  - task:
      name: Audio Classification
      type: audio-classification
    dataset:
      name: GTZAN
      type: marsyas/gtzan
      config: all
      split: train
      args: all
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.77
---

<!-- 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. -->

# distilhubert-finetuned-gtzan

This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7536
- Accuracy: 0.77

## 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: 2e-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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.2844        | 1.0   | 57   | 2.2655          | 0.24     |
| 2.1199        | 2.0   | 114  | 2.0507          | 0.48     |
| 1.7662        | 3.0   | 171  | 1.7611          | 0.56     |
| 1.5841        | 4.0   | 228  | 1.5132          | 0.68     |
| 1.3824        | 5.0   | 285  | 1.3399          | 0.67     |
| 1.2609        | 6.0   | 342  | 1.2268          | 0.72     |
| 1.0785        | 7.0   | 399  | 1.1935          | 0.7      |
| 1.0587        | 8.0   | 456  | 1.0674          | 0.7      |
| 0.9831        | 9.0   | 513  | 0.9904          | 0.74     |
| 0.9783        | 10.0  | 570  | 0.9502          | 0.73     |
| 0.829         | 11.0  | 627  | 0.9090          | 0.76     |
| 0.7314        | 12.0  | 684  | 0.8753          | 0.74     |
| 0.6674        | 13.0  | 741  | 0.8584          | 0.76     |
| 0.8236        | 14.0  | 798  | 0.8069          | 0.78     |
| 0.6861        | 15.0  | 855  | 0.7878          | 0.77     |
| 0.6585        | 16.0  | 912  | 0.7773          | 0.76     |
| 0.5389        | 17.0  | 969  | 0.7695          | 0.78     |
| 0.6257        | 18.0  | 1026 | 0.7907          | 0.76     |
| 0.546         | 19.0  | 1083 | 0.7648          | 0.77     |
| 0.6432        | 20.0  | 1140 | 0.7536          | 0.77     |


### Framework versions

- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0