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: 0.84
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: 0.8438
- Accuracy: 0.84
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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1199 | 1.0 | 113 | 2.0015 | 0.48 |
| 1.4342 | 2.0 | 226 | 1.3339 | 0.62 |
| 1.1096 | 3.0 | 339 | 1.0138 | 0.72 |
| 1.0446 | 4.0 | 452 | 0.8484 | 0.74 |
| 0.6716 | 5.0 | 565 | 0.7406 | 0.75 |
| 0.3906 | 6.0 | 678 | 0.6848 | 0.81 |
| 0.335 | 7.0 | 791 | 0.5893 | 0.81 |
| 0.141 | 8.0 | 904 | 0.7190 | 0.79 |
| 0.1962 | 9.0 | 1017 | 0.7019 | 0.84 |
| 0.0604 | 10.0 | 1130 | 0.7604 | 0.83 |
| 0.0141 | 11.0 | 1243 | 0.8239 | 0.83 |
| 0.0982 | 12.0 | 1356 | 0.7948 | 0.84 |
| 0.0086 | 13.0 | 1469 | 0.8192 | 0.84 |
| 0.0078 | 14.0 | 1582 | 0.8407 | 0.83 |
| 0.0073 | 15.0 | 1695 | 0.8438 | 0.84 |
Framework versions
- Transformers 4.33.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3