🧠 Autism Dataset — Preprocessing, Modeling & Deployment Pipeline

This repository contains a complete, reproducible end-to-end machine learning pipeline for the Autism Spectrum Disorder (ASD) Screening Dataset for Children, covering data preprocessing, exploratory data analysis (EDA), feature engineering, model training, evaluation, and deployment.

The project is implemented using a Jupyter Notebook for experimentation and analysis, and the trained model is deployed using Flask as a lightweight web application.

---

📌 Features

🔧 Data Preprocessing

• Handling missing values
• Encoding categorical variables
• Scaling numerical features
• Outlier detection and removal
• Data leakage prevention using a configurable DROP_LEAKAGE flag
• Finding top 10 features

🤖 Machine Learning Pipeline

• Train/test split
• Classical ML models:
  • Decision Tree Classifer
  • Other baseline classifiers
• Model evaluation using:
  • Accuracy
  • Precision
  • Recall
  • F1-score
• Model & preprocessing pipeline serialization

🌐 Model Deployment (Flask)

• Trained ML model deployed using Flask
• Reuse of saved preprocessing pipeline in production
• Modular and extensible backend structure

---

📁 Project Structure

autism-pipeline/
│
├── autism.csv                       # Cleaned dataset
├── Autism-Child-Data.arff           # Original ARFF dataset
│
├── models.pkl                       # PKL file 
├── preprocessor.pkl                 # PKL file
│
├── autism_pipeline_notebook.ipynb   # EDA + preprocessing + training
│                          
├── templates/
│   └── index.html                   # Fronted
│
├── app.py                           # Flask application
├── train.py                         # Model training & serialization
├── README.md                        # Project documentation
└── .gitignore

---

🚀 Clone the Repository

git clone https://github.com/your-username/autism-pipeline.git
cd autism-pipeline