results

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6945
• Accuracy: 0.4694
• F1score: 0.6
• Recall: 0.8667
• Tn: 7
• Fp: 46
• Fn: 6
• Tp: 39

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-08
• 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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1score	Recall	Tn	Fp	Fn	Tp
No log	1.0	13	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	2.0	26	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	3.0	39	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	4.0	52	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	5.0	65	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	6.0	78	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	7.0	91	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	8.0	104	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	9.0	117	0.6945	0.4694	0.6	0.8667	7	46	6	39
No log	10.0	130	0.6945	0.4694	0.6	0.8667	7	46	6	39

Framework versions

• Transformers 4.42.4
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1