🧠 DEEP LEARNING — Bhavya Kansal

A Live, Production-Grade DL Knowledge Repository

From raw data to trained neural networks — every algorithm, every single neural network properly explained, every experiment, documented.

🏢 Maintained and updated by : Bhavya Kansal  |  🌐 visit at : bhavyakansal.dev  |  📍 Patiala, Punjab, India

---

📋 Table of Contents

• About This Repository
• Who Is This For?
• Tech Stack
• Notebook Index
• Getting Started
• Repository Roadmap
• Datasets
• Acknowledgements
• Contributing
• Legal & License
• Contact

---

🧠 About This Repository

This is not just a notebook dump — it is a structured, continuously updated Deep Learning knowledge base maintained and updated by Bhavya Kansal. As an AI/ML Engineer and Developer, I built this repository to provide Beginner to Advanced level and Structural Understanding of Deep Learning.

Every notebook in this repository:

• Is written from scratch with clean, readable code
• Covers theory + implementation — not just copy-paste code
• Is beginner-friendly — designed so anyone can open it and understand it
• Reflects real internship and coursework experiments, not toy examples

This repository is actively maintained and updated regularly with new neural network architectures, projects, and experiments as the learning journey progresses.

---

👥 Who Is This For?

Audience	How This Helps
🎓 Beginners	Who want to start Deep Learning with clean, documented code
🔬 Students	Reference implementations for assignments and understanding
💼 Practitioners	Quick refresher notebooks for standard architectures
🧑‍💻 Developers	Baseline TensorFlow, ANN, RNN, CNN patterns to build production models from scratch

---

⚙️ Tech Stack

Tool	Purpose
	Core programming language
	Deep learning & neural networks
	High-level neural network API
	Numerical computing & tensors
	Data preprocessing & analysis
	Data visualization
	Statistical visualization
	ML utilities & preprocessing
	Experimentation & prototyping

---

📓 Notebook Index

All notebooks are self-contained and can be opened directly on GitHub or run locally. Click any notebook name to open it.

---

📦 Data Preprocessing & Feature Engineering

The foundation of every DL pipeline — cleaning, transforming, and preparing raw data.

#	Notebook	Concepts Covered
1	Data Preprocessing	Missing values, data cleaning, pipelines
2	Feature Scaling	StandardScaler, MinMaxScaler, normalization
3	Encoding	Label encoding, One-Hot encoding for neural networks
4	Outlier Detection	IQR, Z-score, visualizing outliers

---

🧠 Neural Networks Fundamentals

Building blocks of deep learning — from perceptrons to multi-layer networks.

#	Notebook	Concepts Covered
1	Artifical Neural Network (ANN)	Forward propagation, backpropagation, activations
2	Activation Functions	ReLU, Sigmoid, Tanh, Softmax explained
3	Loss Functions	MSE, Cross-Entropy, choosing right loss
4	Optimizers	SGD, Adam, RMSprop, learning rates

---

🔄 Recurrent Neural Networks (RNN)

For sequences and time-series data.

#	Notebook	Concepts Covered
1	RNN Basics	Vanishing gradients, unfolding, BPTT
2	LSTM Networks	Long Short-Term Memory, cell state, gates
3	GRU Networks	Gated Recurrent Units, simplified LSTM
4	Sequence to Sequence	Encoder-Decoder, attention mechanisms

---

🖼️ Convolutional Neural Networks (CNN)

For image recognition and computer vision tasks.

#	Notebook	Concepts Covered
1	CNN Fundamentals	Convolution, pooling, stride, padding
2	Popular Architectures	LeNet, AlexNet, VGG, ResNet concepts
3	Image Classification	CIFAR-10, MNIST with CNNs
4	Transfer Learning	Fine-tuning pre-trained models

---

🎯 Model Evaluation & Optimization

Because building a model is only half the job.

#	Notebook	Concepts Covered
1	Confusion Matrix	TP/FP/FN/TN, precision, recall, F1
2	Model Validation	K-Fold, Cross-validation strategies
3	Regularization	Dropout, L1/L2, Early Stopping
4	Hyperparameter Tuning	GridSearchCV, RandomizedSearchCV

---

🚀 Getting Started

Prerequisites

Make sure you have Python 3.x installed. Then install the required libraries:

pip install numpy pandas matplotlib seaborn scikit-learn tensorflow jupyter

Running Locally

# 1. Clone this repository
git clone https://github.com/BhavyaKansal20/Deep-Learning.git

# 2. Navigate into the folder
cd Deep-Learning

# 3. Launch Jupyter Notebook
jupyter notebook

Then open any .ipynb file from the Jupyter interface in your browser.

Running on Google Colab

Click the badge below or open any notebook on GitHub and change the URL domain from github.com to colab.research.google.com/github:

---

🗺️ Repository Roadmap

This repository is actively growing. Upcoming additions:

• Advanced CNN Architectures (EfficientNet, MobileNet, Vision Transformers)
• Natural Language Processing (BERT, GPT fundamentals)
• Generative Models (GANs, VAEs, Diffusion Models)
• Reinforcement Learning Basics
• Object Detection (YOLO, Faster R-CNN)
• Semantic Segmentation
• End-to-End DL Projects with real-world datasets
• Model deployment notebooks (TensorFlow Serving, FastAPI)

⭐ Star the repo to get notified when new notebooks are added!

---

📂 Datasets

Datasets used in these notebooks are maintained in a separate dedicated repository to keep this repo clean and lightweight.

🔗 Dataset Repository: Datasets

Some notebooks use built-in datasets (CIFAR-10, MNIST, etc.) which require no external download.

---

🤝 Contributing

Contributions, improvements, and suggestions are warmly welcome!

How to contribute:

1. Fork this repository
2. Create a new branch: git checkout -b feature/your-topic
3. Add your notebook or improvement
4. Commit your changes: git commit -m "Add: Transformer notebook"
5. Push to your branch: git push origin feature/your-topic
6. Open a Pull Request with a clear description

Please read the CONTRIBUTING.md and CODE_OF_CONDUCT.md before submitting.

---

⚖️ Legal & License

MIT License

MIT License

Copyright (c) 2026 Bhavya Kansal

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

See the full LICENSE file.

---

📚 Educational Use Disclaimer

All notebooks and code in this repository are intended strictly for educational and learning purposes. The implementations are for conceptual clarity and skill development, not production deployment without thorough validation.

---

📂 Dataset Attribution

Datasets used across these notebooks may be sourced from:

• Custom datasets of Bhavya Kansal
• TensorFlow built-in datasets (Apache License)
• Scikit-learn built-in datasets (BSD License)
• UCI Machine Learning Repository (varies per dataset)
• Publicly available open-source data

Refer to individual notebooks for specific dataset sources and their respective licenses. All will be checked from this repository: Datasets

---

🔐 Security Policy

For responsible disclosure of any security concerns, please refer to the SECURITY.md file.

---

📬 Contact

Bhavya Kansal | AI/ML Developer | Researcher & Collaborator | जय श्री राम 🙏❤️

📍 Patiala, Punjab, India

---

If this repository helped you learn something new — leave a ⭐

It keeps this project alive and motivates more content to be added.

© 2026 Bhavya Kansal · All Rights Reserved

Built with ❤️ & 🧠 in Patiala, Punjab, India