The following repository explores the underlying concepts of efficient Deep Learning principles applied over CNN.
This repository contains a Jupyter notebook that delves into the fundamentals and applications of Convolutional Neural Networks (CNNs) in deep learning. It is designed to provide a hands-on approach to understanding how CNNs can be used for image recognition tasks.
The "DL_CNN" notebook explores the following key topics:
- Basics of CNN architecture
- Implementing CNNs using popular deep learning libraries
- Training CNNs on image datasets - ImageNet: NaturalImageNet
- Evaluating model performance - ResNet-18
- Hyperparameter Optimisation strategy using Baysian Optimisation - Optuna
- Application examples of CNNs in real-world scenarios
To run the notebook, you will need Jupyter Lab or Jupyter Notebook installed on your system. Additionally, the notebook requires the following Python libraries:
- PyTorch
- NumPy
- Matplotlib
You can install these dependencies using pip:
pip install jupyterlab numpy matplotlib torch torchvision
To start Jupyter Lab and open the notebook, run:
jupyter lab DL_CNN.ipynb
Or, for Jupyter Notebook:
jupyter notebook DL_CNN.ipynb
Follow the instructions within the notebook to explore different CNN models and their applications.
Contributions to the notebook are welcome! Please feel free to fork the repository, make your changes, and submit a pull request.
This project is licensed under the MIT License - see the LICENSE.md file for details.