distilhubert-finetuned-babycry-v6

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: 0.6676
• Accuracy: {'accuracy': 0.8260869565217391}
• F1: 0.7474
• Precision: 0.6824
• Recall: 0.8261

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.7816	2.1739	25	0.7361	{'accuracy': 0.8260869565217391}	0.7474	0.6824	0.8261
0.7056	4.3478	50	0.6957	{'accuracy': 0.8260869565217391}	0.7474	0.6824	0.8261
0.6654	6.5217	75	0.6683	{'accuracy': 0.8260869565217391}	0.7474	0.6824	0.8261

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1