AutoML - Automated Machine Learning Platform

An intelligent, cloud-deployed machine learning platform that automates model selection and training for classification, regression, and computer vision tasks. Features both REST API and web UI interfaces.

Live Demo: https://automl.kahramankaya.com/docs#/

---

✨ Features

• Three ML Task Types:

  • 📊 Tabular Classification - Automatic model selection from Random Forest, SVM, Decision Tree
  • 📈 Regression - Automatic selection from Random Forest, SVR, Decision Tree
  • 🖼️ Vision Classification - Deep learning models (ResNet18, VGG16, MobileNet v2) with transfer learning

• Automatic Model Selection: Trains multiple models and automatically selects the best performer

• GPU Support: CUDA-enabled Docker container for fast training

• Dual Interfaces: FastAPI REST endpoints + Streamlit web UI

• Cloud Ready: Fully containerized and deployed on cloud infrastructure

• Flexible Input: Supports CSV files and image directories

• Pre-configured Transforms: Optimized image preprocessing with augmentation

---

Quick Start

Installation

Clone the repository:

git clone https://github.com/KHRMNKY/AutoMl.git
cd AutoMl

Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

Install dependencies:

pip install -r requirements.txt

---

📚 Usage

Option 1: REST API

Start the FastAPI server:

uvicorn api:app --reload

The API will be available at http://127.0.0.1:8000

View interactive API documentation at http://127.0.0.1:8000/docs

Option 2: Streamlit Web UI

Run the Streamlit interface:

streamlit run streamlit4.py

Access the UI at http://localhost:8501

---

📖 API Endpoints

Vision Classification

• POST /vision_classification/train - Train vision model

  {
    "train_folder": "/path/to/train/folder",
    "learning_rate": 0.001,
    "epochs": 10
  }

• POST /vision_classification/predict - Predict on images

  {
    "image_path": "/path/to/image/folder"
  }

Classification (CSV)

• POST /classification/train - Train classification model

  {
    "train_csv_path": "/path/to/train.csv"
  }

• POST /classification/predict - Get predictions

  {
    "test_csv_path": "/path/to/test.csv"
  }

Regression (CSV)

• POST /regression/train - Train regression model

  {
    "train_csv_path": "/path/to/train.csv"
  }

• POST /regression/predict - Get predictions

  {
    "test_csv_path": "/path/to/test.csv"
  }

---

📁 Project Structure

AutoMl/
├── api.py                    # FastAPI backend server
├── autoML.py                 # Core AutoML class with model logic
├── streamlit4.py             # Streamlit web UI
├── requirements.txt          # Python dependencies
├── Dockerfile                # Docker containerization
├── LICENSE                   # Apache 2.0 License
├── README.md                 # This file
└── data/
    ├── Iris_train.csv        # Sample training data
    ├── Iris_test.csv         # Sample testing data
    └── images/               # Sample image datasets
        ├── apple_pie/
        ├── baby_back_ribs/
        └── baklava/

---

🔧 Configuration

Vision Classification Parameters

• Input Size: 224×224 pixels
• Batch Size: 32
• Train/Test Split: 80/20
• Augmentation: TrivialAugmentWide with 31 magnitude bins
• Normalization: ImageNet standard (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])

Supported Models

Classification/Regression (Tabular):

• Random Forest
• Support Vector Machine (SVM/SVR)
• Decision Tree

Vision Classification:

• ResNet18 (ImageNet pre-trained)
• VGG16 (ImageNet pre-trained)
• MobileNet v2 (ImageNet pre-trained)

---

🐳 Docker Deployment

Build the Docker image:

docker build -t automl:latest .

Run the container:

docker run -p 8081:8081 --gpus all automl:latest

The API will be available at http://localhost:8081

---

📊 Example Usage

Using Python

from autoML import AutoML

# Initialize AutoML
automl = AutoML()

# Train classification model
automl.model_fit('data/Iris_train.csv', task='classification')

# Make predictions
predictions = automl.predict('data/Iris_test.csv')
print(f"Best Model: {automl.model_name}")
print(f"Score: {automl.score}")
print(f"Predictions: {predictions}")

---

📦 Dependencies

• FastAPI: Web framework for building APIs
• PyTorch: Deep learning framework
• Scikit-learn: Machine learning algorithms
• Streamlit: Web UI framework
• Pandas: Data manipulation
• NumPy: Numerical computing
• Torchvision: Computer vision utilities
• Pillow: Image processing

See requirements.txt for complete list with versions.

---

🌐 Deployment

This project is currently deployed at: https://automl.kahramankaya.com/docs#/

The API is hosted on cloud infrastructure with GPU support for accelerated training.

---

📝 Data Format

For Classification/Regression

• Format: CSV files
• Target: Last column should be the target variable
• Features: All other columns are treated as features
• Preprocessing: Automatic handling of categorical variables and feature scaling

For Vision Classification

• Format: Folder structure with class subfolders
• Example:

  train_data/
  ├── class1/
  │   ├── image1.jpg
  │   └── image2.jpg
  └── class2/
      ├── image3.jpg
      └── image4.jpg
  

---