| --- |
| pipeline_tag: text-classification |
| tags: |
| - transformers |
| - text-classification |
| - polarops |
| --- |
| |
| # PolarOps: DistilBERT for Political Polarity Classification |
|
|
| This is a fine-tuned [DistilBERT](https://huggingface.co/distilbert-base-uncased) model for binary text classification on political polarization data. It predicts whether a given sentence is *polarized* or *healthy* based on training data from the PolarOps project. |
|
|
| ## Example Usage |
|
|
| ```python |
| from transformers import pipeline |
| classifier = pipeline("text-classification", model="divilian/polarops") |
| classifier("The government should be overthrown.") |
| ``` |
|
|
| ## Labels |
|
|
| - `healthy` — Civil, constructive language |
| - `polarized` — Toxic or partisan rhetoric |
|
|
| ## Training Details |
|
|
| Trained on X samples using `Trainer()` for Y epochs with learning rate Z. |
|
|
| ## Intended Use |
|
|
| Designed for research and experimentation in political discourse classification. Not suitable for deployment in high-stakes settings. |
|
|
| ## Limitations |
|
|
| - Binary labels only |
| - English language only |
| - May reflect training data biases |
|
|
| ## Author |
|
|
| Stephen Davies ([@divilian](https://huggingface.co/divilian)) |
