|
--- |
|
datasets: |
|
- sem_eval_2018_task_1 |
|
language: |
|
- ar |
|
metrics: |
|
- accuracy |
|
library_name: transformers |
|
pipeline_tag: text-classification |
|
tags: |
|
- transformers |
|
- emotion |
|
- BERT |
|
--- |
|
# Model Card for Model ID |
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
This model aims at classifying arabic text into its corresponding set of emotions by fine-tuning a variant of BERT pre-trained on arabic twitter data. |
|
|
|
### Model Sources [optional] |
|
|
|
<!-- Provide the basic links for the model. --> |
|
|
|
- **Repository:** https://github.com/TomoyeSaa/Emotion-Detection-for-Arabic-Language |
|
|
|
## Evaluation |
|
|
|
<!-- This section describes the evaluation protocols and provides the results. --> |
|
Jaccard accuracy is the main eval metric and the model achieved 56.12% |