metadata
license: apache-2.0
base_model: facebook/wav2vec2-base-960h
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: wav2vec2-base-960h-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: 0.73
wav2vec2-base-960h-finetuned-gtzan
This model is a fine-tuned version of facebook/wav2vec2-base-960h on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 1.0690
- Accuracy: 0.73
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- 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
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.3011 | 0.9956 | 56 | 2.2915 | 0.1 |
| 2.2365 | 1.9911 | 112 | 2.1198 | 0.37 |
| 1.9162 | 2.9867 | 168 | 1.9024 | 0.42 |
| 1.7154 | 4.0 | 225 | 1.7397 | 0.39 |
| 1.757 | 4.9956 | 281 | 1.5732 | 0.47 |
| 1.546 | 5.9911 | 337 | 1.5172 | 0.47 |
| 1.5738 | 6.9867 | 393 | 1.3950 | 0.54 |
| 1.2893 | 8.0 | 450 | 1.4202 | 0.56 |
| 1.2745 | 8.9956 | 506 | 1.2819 | 0.59 |
| 1.2632 | 9.9911 | 562 | 1.2788 | 0.66 |
| 1.2195 | 10.9867 | 618 | 1.1909 | 0.63 |
| 1.1151 | 12.0 | 675 | 1.1605 | 0.62 |
| 1.0165 | 12.9956 | 731 | 1.1202 | 0.67 |
| 0.9418 | 13.9911 | 787 | 1.0747 | 0.73 |
| 0.9686 | 14.9333 | 840 | 1.0690 | 0.73 |
Framework versions
- Transformers 4.43.2
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1