Hubert-fine-tuned-persian

This model is a fine-tuned version of m3hrdadfi/hubert-base-persian-speech-emotion-recognition on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7102
• Accuracy: 0.7585
• Precision: 0.7838
• Recall: 0.6493
• F1: 0.7102
• Precision Neutral: 0.7432
• Recall Neutral: 0.85
• F1 Neutral: 0.7930
• Precision Anger: 0.7838
• Recall Anger: 0.6493
• F1 Anger: 0.7102

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
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 500
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Precision Neutral	Recall Neutral	F1 Neutral	Precision Anger	Recall Anger	F1 Anger
0.6917	1.0	294	0.6704	0.5578	0.5078	0.9701	0.6667	0.8947	0.2125	0.3434	0.5078	0.9701	0.6667
0.6453	2.0	588	0.6077	0.7007	0.8194	0.4403	0.5728	0.6622	0.9187	0.7696	0.8194	0.4403	0.5728
0.5272	3.0	882	0.7842	0.6633	0.9487	0.2761	0.4277	0.6196	0.9875	0.7614	0.9487	0.2761	0.4277
0.4118	4.0	1176	0.6566	0.7449	0.736	0.6866	0.7104	0.7515	0.7937	0.7720	0.736	0.6866	0.7104
0.4956	5.0	1470	0.7102	0.7585	0.7838	0.6493	0.7102	0.7432	0.85	0.7930	0.7838	0.6493	0.7102

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 2.18.0
• Tokenizers 0.21.0