wav2vec-bert-ser-standard

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

• Loss: 2.3597
• F1: 0.5549
• Accuracy: 0.564

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy
30.8905	1.0	16	3.6367	0.1464	0.24
28.7614	2.0	32	3.5061	0.1679	0.256
27.0469	3.0	48	3.3160	0.3390	0.388
27.3445	4.0	64	3.0776	0.3525	0.396
24.3884	5.0	80	2.9147	0.4089	0.452
24.4721	6.0	96	2.7240	0.4445	0.472
22.5651	7.0	112	2.6093	0.5077	0.532
21.9695	8.0	128	2.6026	0.4392	0.476
21.3548	9.0	144	2.3849	0.5656	0.584
18.9157	10.0	160	2.3597	0.5549	0.564

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2