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: 0.77
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: 0.9423
- 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: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 3000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1574 | 8.85 | 500 | 1.8008 | 0.66 |
| 1.5882 | 17.7 | 1000 | 1.3509 | 0.7 |
| 1.2416 | 26.55 | 1500 | 1.1347 | 0.72 |
| 1.037 | 35.4 | 2000 | 1.0163 | 0.74 |
| 0.9152 | 44.25 | 2500 | 0.9583 | 0.76 |
| 0.8556 | 53.1 | 3000 | 0.9423 | 0.77 |
Framework versions
- Transformers 4.32.0
- Pytorch 1.12.1+cu113
- Datasets 2.14.4
- Tokenizers 0.13.3