DL_Projects hands-on deep learning experiments

This repository collects compact, runnable deep-learning projects I built while learning and experimenting with neural networks. Projects are intentionally small so you can run them quickly and learn the core ideas without heavy infrastructure.

What you'll find here

• Practical experiments: image classification, generative models, sequence models, and transfer-learning demos.
• Clean notebooks and scripts that show the full flow: data -> model -> training -> quick evaluation.
• Most projects are runnable on a CPU for small datasets; a GPU is recommended for larger experiments.

Quick start

1. Clone:

   git clone https://github.com/vijaykalore/DL_Projects.git

2. Create and activate a virtual environment (PowerShell):

   python -m venv .venv .\.venv\Scripts\Activate.ps1

3. Install dependencies for a typical project:

   pip install -r requirements.txt

4. Open the project you want and run the main notebook or script:

   jupyter lab

Project index (open the folder name in this repo)

• Project 01 ( Breast Cance Classification with NN ) breast cancer classification with a simple dense NN.
• Project 02 ( MNIST Digit classification using NN ) baseline dense network on MNIST digits.
• Project 03 ( Dog vs Cat Classification Transfer Learning ) transfer learning with pretrained CNN backbones.
• Project 04 ( CIFAR 10 Object Recognition using ResNet50 ) ResNet50-based classifier on CIFAR-10.
• Project 05 ( Face Mask Detection using CNN ) real-time face mask detector with CNN.
• Project 06 ( Fashion MNIST end-to-end Project ) EDA and classification on Fashion-MNIST.
• Project 07 ( Plant Disease Prediction CNN Deep Leanring Project ) plant leaf disease classifier.
• Project 08 ( Neural Network using PyTorch Breast Cancer Prediction ) PyTorch implementation for breast cancer.
• Project 09 ( Generate Handwritten Digit Images DCGAN ) DCGAN for handwritten digit generation.
• Project 10 ( IMDB reviews Sentiment Analysis LSTM ) LSTM-based sentiment classifier on IMDB.
• Extra Learning Notebooks supplementary experiments and notes.

Notes and recommendations

• Many projects download data at runtime. Check each project's notebook/header for dataset download steps and local paths.
• If a project contains a requirements.txt file, use it to install the exact dependencies for reproducibility.
• For visualization-heavy projects, add a results/ folder with sample outputs (images, GIFs) to the project folder this helps reviewers quickly see results.

Contributing

• Want to contribute? Open an issue describing the change or create a focused pull request. Please include a short README for new projects and a note about how you tested changes.

Enjoy exploring run the notebooks, change hyperparameters, and make things your own.

Vijay