IoT Classification with PyTorch

This project demonstrates a simple IoT classification task using synthetic data with PyTorch. The goal is to classify IoT device data into different categories.

Project Structure

• src/: Contains the Python code for the project.
  • generate_iot_data.py: Generates synthetic IoT data.
  • dataset.py: Contains PyTorch Dataset and DataLoader helpers.
  • model.py: Defines the neural network model.
  • train.py: Contains the training and testing loop.
  • utils.py: Provides utility functions like saving metrics to JSON.
• requirements.txt: Lists the Python dependencies.
• .gitignore: Specifies files and directories to be ignored by git.

Setup

1. Install the required packages:

   pip install -r requirements.txt
   

2. Generate synthetic IoT data:

   python -m src.generate_iot_data
   

3. Train the model and produce results:

   python -m src.train --batch-size 32 --epochs 10 --lr 0.001 --train-ratio 0.8
   

4. Install TypeScript and ts-node:

   npm install
   

5. Display the model results:

   npx ts-node ts/src/results.ts
   

Requirements

• Python 3.x
• PyTorch
• NumPy
• Pandas
• Scikit-learn