distilhubert-finetuned-babycry-v5

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: 0.8441
• Accuracy: {'accuracy': 0.782608695652174}
• F1: 0.6872
• Precision: 0.6125
• Recall: 0.7826

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: 2
• total_train_batch_size: 16
• 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
0.7047	1.0870	25	0.9225	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826
0.6071	2.1739	50	0.9175	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826
0.6525	3.2609	75	0.8866	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826
0.6558	4.3478	100	0.8433	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826
0.5577	5.4348	125	0.8705	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826
0.7055	6.5217	150	0.8323	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826
0.6092	7.6087	175	0.8440	{'accuracy': 0.782608695652174}	0.6872	0.6125	0.7826

Framework versions

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