my_april11_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: 0.0029
• 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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6926	1.0	11	0.7243	0.2857
0.6444	2.0	22	0.6651	0.5714
0.5187	3.0	33	0.5439	0.7619
0.4879	4.0	44	0.4134	0.8571
0.3367	5.0	55	0.3302	0.8571
0.2124	6.0	66	0.1248	1.0
0.2257	7.0	77	0.3286	0.8571
0.1128	8.0	88	0.0331	1.0
0.1117	9.0	99	0.0209	1.0
0.0178	10.0	110	0.0142	1.0
0.0149	11.0	121	0.0106	1.0
0.0119	12.0	132	0.0084	1.0
0.01	13.0	143	0.0070	1.0
0.0086	14.0	154	0.0061	1.0
0.0065	15.0	165	0.0053	1.0
0.0065	16.0	176	0.0048	1.0
0.006	17.0	187	0.0044	1.0
0.0055	18.0	198	0.0040	1.0
0.0052	19.0	209	0.0038	1.0
0.0043	20.0	220	0.0035	1.0
0.0045	21.0	231	0.0034	1.0
0.0044	22.0	242	0.0033	1.0
0.0042	23.0	253	0.0031	1.0
0.0041	24.0	264	0.0030	1.0
0.0036	25.0	275	0.0030	1.0
0.0041	26.0	286	0.0029	1.0
0.0039	27.0	297	0.0029	1.0
0.0035	27.2857	300	0.0029	1.0

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0