---
library_name: transformers
license: cc-by-nc-sa-4.0
base_model: amiriparian/ExHuBERT
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: ExHubert-fine-tuned-persian_v2
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# ExHubert-fine-tuned-persian_v2

This model is a fine-tuned version of [amiriparian/ExHuBERT](https://huggingface.co/amiriparian/ExHuBERT) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5637
- Accuracy: 0.8444
- Precision: 0.8209
- Recall: 0.7483
- F1: 0.7829
- Precision Neutral: 0.8566
- Recall Neutral: 0.9020
- F1 Neutral: 0.8787
- Precision Anger: 0.8209
- Recall Anger: 0.7483
- F1 Anger: 0.7829

## 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: 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
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     | Precision Neutral | Recall Neutral | F1 Neutral | Precision Anger | Recall Anger | F1 Anger |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:-----------------:|:--------------:|:----------:|:---------------:|:------------:|:--------:|
| 0.8717        | 1.0   | 393  | 0.7755          | 0.6454   | 0.5145    | 0.9660 | 0.6714 | 0.9569            | 0.4531         | 0.6150     | 0.5145          | 0.9660       | 0.6714   |
| 0.4644        | 2.0   | 786  | 0.4577          | 0.8495   | 0.7785    | 0.8367 | 0.8066 | 0.8974            | 0.8571         | 0.8768     | 0.7785          | 0.8367       | 0.8066   |
| 0.4535        | 3.0   | 1179 | 0.4818          | 0.8546   | 0.8       | 0.8163 | 0.8081 | 0.8884            | 0.8776         | 0.8830     | 0.8             | 0.8163       | 0.8081   |
| 0.4773        | 4.0   | 1572 | 0.5514          | 0.8189   | 0.7289    | 0.8231 | 0.7732 | 0.8850            | 0.8163         | 0.8493     | 0.7289          | 0.8231       | 0.7732   |
| 0.3337        | 5.0   | 1965 | 0.5680          | 0.8112   | 0.7417    | 0.7619 | 0.7517 | 0.8548            | 0.8408         | 0.8477     | 0.7417          | 0.7619       | 0.7517   |
| 0.2774        | 6.0   | 2358 | 0.6004          | 0.8367   | 0.8879    | 0.6463 | 0.7480 | 0.8175            | 0.9510         | 0.8792     | 0.8879          | 0.6463       | 0.7480   |
| 0.2007        | 7.0   | 2751 | 0.5529          | 0.8546   | 0.8629    | 0.7279 | 0.7897 | 0.8507            | 0.9306         | 0.8889     | 0.8629          | 0.7279       | 0.7897   |
| 0.2025        | 8.0   | 3144 | 0.5655          | 0.8444   | 0.8162    | 0.7551 | 0.7845 | 0.8594            | 0.8980         | 0.8782     | 0.8162          | 0.7551       | 0.7845   |
| 0.2583        | 9.0   | 3537 | 0.5635          | 0.8444   | 0.8258    | 0.7415 | 0.7814 | 0.8538            | 0.9061         | 0.8792     | 0.8258          | 0.7415       | 0.7814   |
| 0.3264        | 10.0  | 3930 | 0.5637          | 0.8444   | 0.8209    | 0.7483 | 0.7829 | 0.8566            | 0.9020         | 0.8787     | 0.8209          | 0.7483       | 0.7829   |


### Framework versions

- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0