Deep Learning Hands-On Practice Repository

This repository contains my end-to-end hands-on Deep Learning learning journey, where I implemented core deep learning concepts from scratch and using Keras/TensorFlow.
Each notebook focuses on a single concept, building intuition through experiments, visualizations, and practical implementations.

The goal of this repository is to create a structured, topic-wise reference for deep learning fundamentals and advanced neural network architectures.

---

What This Repository Covers

This repository includes practical implementations of:

• Artificial Neural Networks (ANN)
• Optimization Techniques
• Regularization Methods
• Convolutional Neural Networks (CNN)
• Transfer Learning
• Recurrent Neural Networks (RNN)
• NLP preprocessing (Tokenization, Encoding)
• Hyperparameter Tuning using Keras Tuner

Each topic is implemented as an independent Jupyter Notebook for clarity and ease of understanding.

---

Skills Demonstrated

• Designing and training Artificial Neural Networks (ANN)
• Implementing Gradient Descent & Optimization techniques
• Handling Vanishing/Exploding Gradients
• Applying Regularization techniques (Dropout, L1/L2, BatchNorm)
• Building Convolutional Neural Networks (CNN) from scratch
• Working with Pretrained Models & Transfer Learning
• Implementing Recurrent Neural Networks (RNN) for NLP tasks
• Text preprocessing using Tokenization & Integer Encoding
• Hyperparameter optimization using Keras Tuner
• Model evaluation and performance tuning

---

Topic-wise Breakdown

Neural Network Fundamentals

• Perceptron Introduction & Tricks
• Gradient Descent Optimization
• Vanishing Gradient Problem
• Feature Scaling

Regularization & Optimization

• Dropout (Classification & Regression)
• L1 / L2 Regularization
• Batch Normalization
• Early Stopping

Artificial Neural Networks (ANN)

• Customer Churn Prediction using ANN
• Dense Neural Network Design & Training

Convolutional Neural Networks (CNN)

• Padding & Strides
• Pooling Layers
• LeNet Architecture
• Pretrained CNN Models

Transfer Learning

• Cat vs Dog Image Classification
• Fine-tuning Pretrained Models
• Feature Extraction Techniques

Recurrent Neural Networks (RNN)

• RNN Architecture Basics
• Integer Encoding & Tokenization
• Sentiment Classification using IMDB Dataset

Hyperparameter Tuning

• Keras Tuner for Automated Hyperparameter Optimization

Author

Narendra Tekale
Aspiring Data Scientist | Machine Learning & Deep Learning Enthusiast

Portfolio: https://narendratekale.com/

Acknowledgements

A big thank you to CampusX (Nitish Singh) for creating such an excellent and well-structured Deep Learning playlist on YouTube.

The content played a significant role in building my conceptual understanding and practical skills in:

• Neural Networks fundamentals
• CNN, RNN, and Transfer Learning
• Model optimization and regularization techniques

This repository is inspired by the learnings from that playlist and represents my hands-on practice and independent experimentation based on those concepts.