Metadata-Version: 2.4
Name: ml-etl-package
Version: 0.1.0
Summary: A simple ML pipeline package with ETL steps using scikit-learn
Author-email: Your Name <you@example.com>
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: pandas
Requires-Dist: scikit-learn
Requires-Dist: joblib

# README.md
# ML ETL Package

A lightweight Python package for running a simple Machine Learning ETL pipeline using Scikit-learn.

## Features
- 📥 Extract: Load data from CSV files
- 🔧 Transform: Clean and scale data with `StandardScaler`
- 📤 Load: Save transformed data to CSV
- 🤖 Model: Train and save a Linear Regression model
- 🔮 Predict: Make predictions from scaled inputs

## Installation
```bash
pip install ml-etl-package
```

## Usage Example
```python
from ml_etl_package.extract import extract_from_csv
from ml_etl_package.transform import transform
from ml_etl_package.load import save_to_csv
from ml_etl_package.model import train_and_save_model, load_model
from ml_etl_package.predict import predict

# Extract
df = extract_from_csv("data.csv")

# Transform
X, y, scaler = transform(df, feature_cols=["sqft", "bedrooms", "bathrooms"], target_col="price")

# Load (save transformed data)
save_to_csv(X, y, filename="transformed_output.csv")

# Train
train_and_save_model(X, y)
model = load_model()

# Predict
input_data = [1200, 3, 2]  # sqft, bedrooms, bathrooms
result = predict(model, scaler, input_data)
print(f"Predicted price: {result:.2f}")
```

## License
MIT
