---
base_model: aubmindlab/bert-base-arabertv02
tags:
- generated_from_trainer
datasets:
- offenseval_2020
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: ArabertHateSpeech
  results:
  - task:
      name: Text Classification
      type: text-classification
    dataset:
      name: offenseval_2020
      type: offenseval_2020
      config: ar
      split: test
      args: ar
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.9425287356321839
    - name: F1
      type: f1
      value: 0.8543689320388349
    - name: Precision
      type: precision
      value: 0.875
    - name: Recall
      type: recall
      value: 0.8346883468834688
---

<!-- 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. -->

# ArabertHateSpeech

This model is a fine-tuned version of [aubmindlab/bert-base-arabertv02](https://huggingface.co/aubmindlab/bert-base-arabertv02) on the offenseval_2020 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2500
- Accuracy: 0.9425
- F1: 0.8544
- Precision: 0.875
- Recall: 0.8347

## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
| No log        | 1.0   | 490  | 0.1377          | 0.9321   | 0.8263 | 0.8551    | 0.7995 |
| 0.1418        | 2.0   | 980  | 0.0967          | 0.9321   | 0.8121 | 0.9210    | 0.7263 |
| 0.0898        | 3.0   | 1470 | 0.1082          | 0.9442   | 0.8517 | 0.9185    | 0.7940 |
| 0.0595        | 4.0   | 1960 | 0.1530          | 0.9338   | 0.8358 | 0.8370    | 0.8347 |
| 0.0405        | 5.0   | 2450 | 0.1559          | 0.9442   | 0.8579 | 0.8825    | 0.8347 |
| 0.0194        | 6.0   | 2940 | 0.2175          | 0.9398   | 0.8541 | 0.8364    | 0.8726 |
| 0.0153        | 7.0   | 3430 | 0.1994          | 0.9392   | 0.8452 | 0.8707    | 0.8211 |
| 0.0102        | 8.0   | 3920 | 0.2154          | 0.9403   | 0.8541 | 0.8439    | 0.8645 |
| 0.0093        | 9.0   | 4410 | 0.2296          | 0.9409   | 0.8470 | 0.8872    | 0.8103 |
| 0.0047        | 10.0  | 4900 | 0.2406          | 0.9420   | 0.8524 | 0.8768    | 0.8293 |
| 0.0038        | 11.0  | 5390 | 0.2530          | 0.9436   | 0.8591 | 0.8674    | 0.8509 |
| 0.0051        | 12.0  | 5880 | 0.2500          | 0.9425   | 0.8544 | 0.875     | 0.8347 |


### Framework versions

- Transformers 4.33.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3