metadata
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
- generated_from_trainer
datasets:
- arisha/stuttering
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-stutteringdetection
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: stuttering
type: arisha/stuttering
metrics:
- name: Accuracy
type: accuracy
value: 0.7692307692307693
distilhubert-finetuned-stutteringdetection
This model is a fine-tuned version of ntu-spml/distilhubert on the stuttering dataset. It achieves the following results on the evaluation set:
- Loss: 0.8952
- Accuracy: 0.7692
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.1755 | 1.0 | 102 | 1.1561 | 0.5275 |
| 0.9759 | 2.0 | 204 | 0.9051 | 0.6703 |
| 0.5208 | 3.0 | 306 | 0.7956 | 0.7143 |
| 0.3765 | 4.0 | 408 | 0.7282 | 0.8022 |
| 0.2368 | 5.0 | 510 | 0.6921 | 0.8022 |
| 0.1761 | 6.0 | 612 | 0.8270 | 0.7582 |
| 0.3561 | 7.0 | 714 | 0.8967 | 0.7253 |
| 0.2222 | 8.0 | 816 | 0.8201 | 0.8022 |
| 0.0303 | 9.0 | 918 | 0.9433 | 0.7473 |
| 0.019 | 10.0 | 1020 | 0.8952 | 0.7692 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1