---
language: en
tags:
  - sklearn
  - diabetes
  - regression
  - gradient-boosting
---

# Diabetes Disease Progression Predictor

A **Gradient Boosting Regressor** trained on the
[sklearn Diabetes dataset](https://scikit-learn.org/stable/datasets/toy_dataset.html#diabetes-dataset)
to predict disease progression one year after baseline (continuous target).

## Results

| Metric | Value |
|--------|-------|
| Test MAE | **44.80** |
| Test R² | **0.4366** |
| CV R² (5-fold) | **0.3543** |

## Feature Importances

| Feature | Importance |
|---------|-----------|
| bmi | 0.3688 |
| s5 | 0.2360 |
| bp | 0.0869 |
| s2 | 0.0763 |
| age | 0.0559 |
| s6 | 0.0495 |
| s1 | 0.0474 |
| s3 | 0.0392 |
| s4 | 0.0277 |
| sex | 0.0122 |

## Features

The model uses 10 baseline variables: age, sex, BMI, blood pressure,
and 6 blood serum measurements (s1–s6).

## Usage

```python
import joblib
from huggingface_hub import hf_hub_download

model_path = hf_hub_download(repo_id="Soulay/diabetes-predictor", filename="model.joblib")
model = joblib.load(model_path)

# [age, sex, bmi, bp, s1, s2, s3, s4, s5, s6]  (standardised values)
prediction = model.predict([[0.05, -0.04, 0.06, 0.02, -0.01, 0.0, -0.03, 0.04, 0.02, -0.01]])
print(prediction)  # e.g. [152.3]
```

## Training

Model trained automatically via GitHub Actions CI/CD on every push to `main`.