metadata

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: 0.81

distilhubert-finetuned-gtzan

This model is a fine-tuned version of ntu-spml/distilhubert on the GTZAN dataset. It achieves the following results on the evaluation set:

Loss: 1.1842
Accuracy: 0.81

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: 0.0001
train_batch_size: 32
eval_batch_size: 32
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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1069	1.0	29	2.0003	0.46
1.8026	2.0	58	1.6073	0.59
1.3938	3.0	87	1.2140	0.72
1.0295	4.0	116	1.0740	0.64
0.8339	5.0	145	0.9243	0.71
0.6347	6.0	174	0.8837	0.72
0.4137	7.0	203	0.8274	0.78
0.3162	8.0	232	0.7596	0.82
0.2055	9.0	261	0.8541	0.77
0.2237	10.0	290	0.7220	0.78
0.0601	11.0	319	0.7765	0.81
0.0817	12.0	348	0.7603	0.86
0.0196	13.0	377	0.8611	0.8
0.0641	14.0	406	0.9281	0.8
0.0253	15.0	435	1.2051	0.77
0.0079	16.0	464	1.1073	0.81
0.0055	17.0	493	1.0920	0.81
0.012	18.0	522	1.1882	0.82
0.0051	19.0	551	1.0023	0.81
0.0047	20.0	580	1.2339	0.79
0.0036	21.0	609	1.1471	0.79
0.0033	22.0	638	1.1924	0.8
0.0032	23.0	667	1.1064	0.81
0.0028	24.0	696	1.1140	0.8
0.0026	25.0	725	1.1344	0.81
0.0163	26.0	754	1.1551	0.8
0.0027	27.0	783	1.1843	0.81
0.0025	28.0	812	1.1824	0.81
0.0104	29.0	841	1.1636	0.8
0.0047	30.0	870	1.1842	0.81

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu118
Datasets 2.14.0
Tokenizers 0.13.3