|
--- |
|
license: apache-2.0 |
|
tags: |
|
- setfit |
|
- sentence-transformers |
|
- text-classification |
|
pipeline_tag: text-classification |
|
--- |
|
|
|
# /content/model |
|
|
|
This model is used for binary classification of racially/culturally exclusive or exclusive sentences in French and English languages. |
|
It is a fine-tuned model of [morit/french_xlm_xnli](https://huggingface.co/morit/french_xlm_xnli) |
|
on [this dataset](https://github.com/BeToast/Racially-Exclusive-Speech-Detection/tree/main/final/data_ready) |
|
with [SetFit model](https://github.com/huggingface/setfit) |
|
|
|
This model generalizes exclusivity well which is shown by the ability to detect exclusivity even to fictional races. |
|
All code for entire experiments is in my [gitrepo](https://github.com/BeToast/Racially-Exclusive-Speech-Detection) |
|
|
|
## Usage |
|
|
|
##### Use example [Google Collab notebook](https://colab.research.google.com/drive/1EABmyXsjQihRS1W9If-weg-lR8i4xn_4?usp=sharing) |
|
|
|
##### Use this model in your own code: |
|
```bash |
|
python -m pip install setfit |
|
``` |
|
|
|
You can then run predictions with the model: |
|
|
|
```python |
|
from setfit import SetFitModel |
|
|
|
# Download from Hub and run inference |
|
model = SetFitModel.from_pretrained("/content/model") |
|
|
|
# Run inclusive or exclusive binary classification. |
|
predictions = model(["Elves are all snobs", "Elves are known for their high intelligence and wealth"]) |
|
predictions |
|
``` |
|
|