mydata-iau

This model is a fine-tuned version of facebook/wav2vec2-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6266
• Accuracy: 0.5385

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7031	1.0	4	0.6858	0.5385
0.631	2.0	8	0.6973	0.3846
0.7272	3.0	12	0.7615	0.3846
0.5345	4.0	16	0.7592	0.4615
0.7417	5.0	20	0.7341	0.4615
0.6045	6.0	24	0.7095	0.4615
0.5344	7.0	28	0.7080	0.3846
0.4422	8.0	32	0.7393	0.4615
0.4156	9.0	36	0.7389	0.4615
0.3876	10.0	40	0.7068	0.5385
0.3761	11.0	44	0.6788	0.5385
0.433	12.0	48	0.6340	0.5385
0.3416	13.0	52	0.6109	0.5385
0.3447	14.0	56	0.6167	0.5385
0.3328	15.0	60	0.6266	0.5385

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0
• Datasets 2.12.0
• Tokenizers 0.15.1