wav2vec2-base-random-stop-classification-1

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

• Loss: 0.4066
• Accuracy: 0.8651

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6949	0.99	18	0.6706	0.5906
0.6753	1.97	36	0.6470	0.6383
0.6231	2.96	54	0.5590	0.7302
0.544	4.0	73	0.4623	0.7977
0.4806	4.99	91	0.4061	0.8317
0.4543	5.97	109	0.5891	0.7643
0.4947	6.96	127	0.3944	0.8386
0.4431	8.0	146	0.4528	0.8093
0.4147	8.99	164	0.4560	0.8222
0.4094	9.97	182	0.4193	0.8447
0.3906	10.96	200	0.3846	0.8549
0.3835	12.0	219	0.3845	0.8569
0.3632	12.99	237	0.3660	0.8644
0.3622	13.97	255	0.4107	0.8617
0.3472	14.96	273	0.3733	0.8685
0.3419	16.0	292	0.4496	0.8467
0.3074	16.99	310	0.3987	0.8638
0.3278	17.97	328	0.3740	0.8665
0.2841	18.96	346	0.3999	0.8651
0.2837	20.0	365	0.3954	0.8604
0.2928	20.99	383	0.3871	0.8644
0.3002	21.97	401	0.4978	0.8386
0.2783	22.96	419	0.4079	0.8692
0.2703	24.0	438	0.3977	0.8713
0.2816	24.66	450	0.4066	0.8651

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.0
• Datasets 2.7.1
• Tokenizers 0.13.2