metadata
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
- generated_from_trainer
datasets:
- audiofolder
metrics:
- accuracy
model-index:
- name: heartbeat-detection
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: audiofolder
type: audiofolder
config: default
split: train[:90]
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.9629629629629629
heartbeat-detection
This model is a fine-tuned version of ntu-spml/distilhubert on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.4766
- Accuracy: 0.9630
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: 1e-06
- 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: 0.01
- num_epochs: 16
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.6241 | 1.0 | 8 | 1.6177 | 0.0 |
| 1.6057 | 2.0 | 16 | 1.5995 | 0.0 |
| 1.5872 | 3.0 | 24 | 1.5829 | 0.0370 |
| 1.5703 | 4.0 | 32 | 1.5674 | 0.0741 |
| 1.5557 | 5.0 | 40 | 1.5532 | 0.2593 |
| 1.5415 | 6.0 | 48 | 1.5401 | 0.4815 |
| 1.5285 | 7.0 | 56 | 1.5282 | 0.7037 |
| 1.5172 | 8.0 | 64 | 1.5175 | 0.7778 |
| 1.5074 | 9.0 | 72 | 1.5080 | 0.8519 |
| 1.4975 | 10.0 | 80 | 1.4998 | 0.8519 |
| 1.4906 | 11.0 | 88 | 1.4928 | 0.9259 |
| 1.4844 | 12.0 | 96 | 1.4870 | 0.9259 |
| 1.4788 | 13.0 | 104 | 1.4825 | 0.9630 |
| 1.4744 | 14.0 | 112 | 1.4793 | 0.9630 |
| 1.4718 | 15.0 | 120 | 1.4773 | 0.9630 |
| 1.4704 | 16.0 | 128 | 1.4766 | 0.9630 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2