wav2vec2-large-xlsr-53-english-finetuned-babycry-v3

This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-english on the audiofolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.7337
• Accuracy: {'accuracy': 0.8152173913043478}
• F1: 0.7322
• Precision: 0.6646
• Recall: 0.8152

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.949	0.5435	25	0.7351	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.7488	1.0870	50	0.7795	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.6911	1.6304	75	0.7066	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.8113	2.1739	100	0.8012	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.634	2.7174	125	0.7801	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.6503	3.2609	150	0.7712	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.7523	3.8043	175	0.7078	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.5493	4.3478	200	0.7484	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152
0.7253	4.8913	225	0.7341	{'accuracy': 0.8152173913043478}	0.7322	0.6646	0.8152

Framework versions

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