Deep Learning Fundamentals: Practice & Implementation This repository contains my journey through Deep Learning. It includes code provided in campusx yt tutorial as well as my own personal implementations, experiments, and notes as I worked through the concepts.

🚀 Overview The goal of this project was to move beyond theoretical knowledge and gain hands-on experience with neural networks, optimization techniques, and various deep learning architectures.

📂 Repository Structure

practice notebooks/: my practice notebooks as well as from tutorial

deep_learning_notes/: Short summaries of the key concepts learned in each module.

🧠 Concepts Covered Throughout this repository, I have implemented and explored:

Neural Network Basics: Perceptrons, activation functions (ReLU, Sigmoid, Softmax).

Backpropagation: Understanding how gradients flow through a network to update weights.

Optimization: Stochastic Gradient Descent (SGD), Adam, and RMSProp.

Loss Functions: Cross-Entropy for classification and MSE for regression.

Architectures:

MLP: Multi-Layer Perceptrons.

CNN: Convolutional Neural Networks for image processing.

RNN/LSTM: For sequential data and time-series analysis.

Transformer Architecture in Detail

🛠️ Tools & Frameworks Language: Python 3.x

Libraries: TensorFlow, NumPy, Pandas, Matplotlib, Scikit-learn.

📈 Key Learnings Hyperparameter Tuning: I learned how sensitive models are to learning rates and batch sizes.

Overfitting: Implemented Dropout and L2 Regularization to improve model generalization.

Data Preprocessing: The importance of normalization and data augmentation in training robust models.

🔗 Credits The base code and educational structure were provided by:

Tutorial Name: CampusX 100 days of deep learning

How to Use Clone the repo: git clone https://github.com/asim-shah-web/deep-learning-practice.git