my_CURRENT_model

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

• Loss: 0.0022
• Accuracy: 1.0

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 35

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6977	1.0	11	0.6867	0.4762
0.6549	2.0	22	0.7040	0.3333
0.582	3.0	33	0.5344	0.7619
0.4289	4.0	44	0.2738	1.0
0.1483	5.0	55	0.1308	1.0
0.1315	6.0	66	1.2406	0.4286
0.1517	7.0	77	0.0407	1.0
0.0996	8.0	88	0.0259	1.0
0.11	9.0	99	0.2798	0.9048
0.0162	10.0	110	0.0135	1.0
0.014	11.0	121	0.0103	1.0
0.0113	12.0	132	0.0083	1.0
0.0095	13.0	143	0.0069	1.0
0.0081	14.0	154	0.0059	1.0
0.006	15.0	165	0.0052	1.0
0.006	16.0	176	0.0046	1.0
0.0054	17.0	187	0.0041	1.0
0.005	18.0	198	0.0038	1.0
0.0046	19.0	209	0.0035	1.0
0.0038	20.0	220	0.0032	1.0
0.0039	21.0	231	0.0030	1.0
0.0037	22.0	242	0.0029	1.0
0.0036	23.0	253	0.0027	1.0
0.0034	24.0	264	0.0026	1.0
0.0029	25.0	275	0.0025	1.0
0.0031	26.0	286	0.0024	1.0
0.0031	27.0	297	0.0024	1.0
0.003	28.0	308	0.0023	1.0
0.0029	29.0	319	0.0023	1.0
0.0026	30.0	330	0.0023	1.0
0.0029	31.0	341	0.0022	1.0
0.0029	31.8571	350	0.0022	1.0

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1