Overview

This repository contains a collection of projects developed using Jupyter Notebooks, focusing on testing various machine learning pipelines and neural network models. Some projects also involve a statistical machine learning approach, showcasing the analysis's versatility and depth.

Projects

• Clustering Algorithms
• DCGAN Model
• Deep Learning with Bayesian Inference
• Distribute Strategy Deep Learning Model
• Knowledge Graph Model
• Low-Rank Factorization
• Regression Models
• Sports Article Classification Model
• Tensorflow Probability Models
• Variety Deep Learning Models

This repository contains implementations of various deep learning and machine learning models and techniques. Below is an overview of the key projects:

BackPropagation Neural Network

Implementation of a neural network trained using backpropagation in Python. The network uses stochastic gradient descent and sigmoid activation functions1.

Clustering Algorithms

Implementations of unsupervised clustering algorithms like K-Means and Mean Shift using different machine learning frameworks including NumPy, PyTorch, TensorFlow, and JAX2.

DCGAN Model

A PyTorch implementation of Deep Convolutional Generative Adversarial Networks (DCGAN) for generating face images. Trained on the CelebA dataset3.

Bayesian Deep Learning

Library for Bayesian neural networks in PyTorch, implementing variational inference, MC-dropout, stochastic gradient MCMC, and Laplace approximation4.

Knowledge Graph Model

Platform for building, exchanging and reusing knowledge graphs. Includes ETL patterns for ingesting data in a standardized format5.

Low-Rank Factorization

Implementing low-rank matrix factorization techniques like Tucker decomposition for compressing convolutional neural networks6. Sports Article Classification Topic modeling approaches (LSA, NMF, LDA) for classifying BBC sports articles into different categories7.

TensorFlow Probability Models

Examples of probabilistic models built using the TensorFlow Probability library, including distributions, variational inference, and MCMC

Getting Started

To explore the projects and run the Jupyter Notebooks locally, follow these steps:

1. Clone the repository: $ git clone https://github.com/yourusername/machine-learning-projects.git
2. Install the required dependencies: $ pip install -r requirements.txt
3. Launch Jupyter Notebook: $ jupyter notebook
4. Open the desired project notebook and run the cells sequentially.

Libraries and Frameworks used in these projects

• Python 3.2
• Jupyter Notebook
• NumPy
• Pandas
• Scikit-learn
• TensorFlow
• Keras
• Matplotlib
• Seaborn

License

This project is licensed under the MIT License - see the LICENSE.md file for details.