100 Days of Machine Learning

This is my journey of learning and applying machine learning algorithms and techniques using Python and PyTorch. I have completed various projects on different topics and datasets, such as classification, regression, clustering, dimensionality reduction, etc. I have also explored some advanced topics, such as deep learning, computer vision, natural language processing, etc. I have learned a lot from this challenge, and I hope to continue learning and improving my skills in machine learning.

You can check out my LinkedIn profile here to see some of my posts and updates on this challenge.

Table of Contents

• Installation
• Usage
• Features
• Technologies
• Acknowledgments

Installation

To install and run this project, you will need Python 3.7 or higher and the following libraries:

• numpy
• pandas
• matplotlib
• scikit-learn
• pytorch
• transformers
• diffusers
• hugging face

You can install these libraries using pip or conda. For example:

pip install numpy pandas matplotlib scikit-learn pytorch

To clone this repository, use the following command:

git clone https://github.com/mayankpuvvala/100DaysOfML.git

To run the notebooks, you can use Jupyter Notebook or Google Colab. For example:

jupyter notebook Abalone_NaiveBayes.ipynb

Usage

This project contains various notebooks that demonstrate different machine learning algorithms and techniques. Each notebook has a title, a description, and a code section. The notebooks also have comments and explanations to help you understand the logic and the results. For example, here is a screenshot of the Abalone_NaiveBayes.ipynb notebook, which shows how to implement Naive Bayes classifier on the Abalone dataset. The goal is to predict the age of abalone age based on their physical measurements.

Features

This project has the following features:

• Implementation of various machine learning algorithms, such as Naive Bayes, Logistic Regression, SVM, Decision Trees, KNN, etc.
• Application of machine learning techniques, such as data preprocessing, feature engineering, model evaluation, hyperparameter tuning, optimization, production, etc.
• Visualization of data and results using matplotlib and seaborn.
• Use of PyTorch, keras as the main framework for deep learning.
• Use of Kaggle for cloud computing and GPU acceleration.

Some of the challenges or limitations of this project are:

• The datasets used are mostly small and simple, which may not reflect the real-world scenarios and complexities.
• The notebooks are not extensively optimized for performance or efficiency, which may affect the speed and accuracy of the models.

Technologies

This project was created with:

• Python 3.11
• NumPy 1.19.5
• Pandas 1.2.4
• Matplotlib 3.4.2
• Scikit-learn 0.24.2
• PyTorch 1.8.1
• Google Colab
• kaggle
• Jupyter Notebooks (mostly)

Acknowledgments

I would like to thank the NPTEL & FastAI sources for providing the datasets, code examples, and tutorials that I used in this project:

• Kaggle for providing various datasets and competitions.
• PyTorch for providing the documentation and tutorials on deep learning.
• Google Colab for providing the cloud platform and GPU support.
• Medium for providing various articles and blogs on machine learning.