metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- audiofolder
metrics:
- accuracy
- f1
model-index:
- name: wav2vec2-base-finetuned-ks
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: audiofolder
type: audiofolder
config: Data_Train
split: train
args: Data_Train
metrics:
- name: Accuracy
type: accuracy
value: 0.6510736196319018
- name: F1
type: f1
value: 0.5657842671346605
wav2vec2-base-finetuned-ks
This model is a fine-tuned version of facebook/wav2vec2-base on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.3216
- Accuracy: 0.6511
- F1: 0.5658
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 2.3065 | 1.0 | 1449 | 2.3713 | 0.2991 | 0.1702 |
| 1.6849 | 2.0 | 2898 | 1.6462 | 0.5560 | 0.4292 |
| 1.4551 | 3.0 | 4347 | 1.3216 | 0.6511 | 0.5658 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3