| title: Credit Worthiness Risk Classification | |
| emoji: "📈" | |
| colorFrom: blue | |
| colorTo: indigo | |
| sdk: gradio | |
| sdk_version: "6.14.0" | |
| python_version: "3.10" | |
| app_file: app.py | |
| pinned: false | |
| # Credit Worthiness Risk Classification | |
| End-to-end classification case study for credit risk prediction, including preprocessing, model selection, and app delivery. | |
| ## Objective | |
| Build and evaluate models to classify credit applicants by risk profile in a reproducible workflow. | |
| ## What Is Included | |
| - data preparation and feature engineering steps; | |
| - supervised models: Logistic Regression, Decision Tree, Random Forest; | |
| - hyperparameter tuning with `GridSearchCV`; | |
| - model evaluation pipeline; | |
| - lightweight Gradio app for interactive inference (`app.py`). | |
| ## Dataset | |
| - Local path: `data/raw/german_credit.csv` | |
| - Source: [Kaggle - German Credit Dataset](https://www.kaggle.com/datasets/mpwolke/cusersmarildownloadsgermancsv) | |
| ## Run Locally | |
| ```bash | |
| pip install -r requirements.txt | |
| python app.py | |
| ``` | |
| ## Project Metadata | |
| - `sdk`: gradio | |
| - `sdk_version`: 6.14.0 | |
| - `app_file`: `app.py` | |
| - `license`: Apache-2.0 | |