wav2vec2-base-random-stop-classification-5

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

• Loss: 0.4239
• Accuracy: 0.8631

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
0.6916	0.99	18	0.6503	0.6362
0.6628	1.97	36	0.5354	0.7391
0.5922	2.96	54	0.4775	0.7786
0.5158	4.0	73	0.4559	0.8072
0.4733	4.99	91	0.4308	0.8188
0.4935	5.97	109	0.5186	0.7888
0.4512	6.96	127	0.4108	0.8358
0.4397	8.0	146	0.4692	0.8270
0.4037	8.99	164	0.4049	0.8304
0.4053	9.97	182	0.4054	0.8379
0.3774	10.96	200	0.4330	0.8379
0.3624	12.0	219	0.3800	0.8495
0.376	12.99	237	0.5123	0.8263
0.3908	13.97	255	0.4049	0.8386
0.3405	14.96	273	0.4200	0.8529
0.3542	16.0	292	0.4040	0.8569
0.3284	16.99	310	0.4578	0.8474
0.3094	17.97	328	0.4465	0.8522
0.2999	18.96	346	0.4126	0.8569
0.3059	20.0	365	0.4139	0.8529
0.2891	20.99	383	0.4101	0.8624
0.2968	21.97	401	0.4589	0.8501
0.2764	22.96	419	0.4263	0.8522
0.2841	24.0	438	0.4350	0.8597
0.2805	24.66	450	0.4239	0.8631

Framework versions

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