metadata

library_name: transformers
license: apache-2.0
base_model: m3hrdadfi/hubert-base-persian-speech-emotion-recognition
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: Hubert-fine-tuned-persian
    results: []

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: 0.5445
Accuracy: 0.7483
Precision: 0.7
Recall: 0.7836
F1: 0.7394
Precision Neutral: 0.7986
Recall Neutral: 0.7188
F1 Neutral: 0.7566
Precision Anger: 0.7
Recall Anger: 0.7836
F1 Anger: 0.7394

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: 2.5e-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: reduce_lr_on_plateau
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
0.7337	1.0	294	0.6602	0.5816	0.5216	0.9925	0.6838	0.9744	0.2375	0.3819	0.5216	0.9925	0.6838
0.5922	2.0	588	0.5445	0.7483	0.7	0.7836	0.7394	0.7986	0.7188	0.7566	0.7	0.7836	0.7394
0.4774	3.0	882	0.7353	0.7177	0.7257	0.6119	0.6640	0.7127	0.8063	0.7566	0.7257	0.6119	0.6640
0.4527	4.0	1176	0.6275	0.7143	0.6582	0.7761	0.7123	0.7794	0.6625	0.7162	0.6582	0.7761	0.7123
0.5922	5.0	1470	0.8464	0.7347	0.6647	0.8433	0.7434	0.8306	0.6438	0.7254	0.6647	0.8433	0.7434

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu124
Datasets 2.18.0
Tokenizers 0.21.0