README.md

Practical Machine Learning and Deep Learning Roadmap

1. Programming Basics

• Learn Python programming
  • Python Tutorial (w3schools.com)
  • Legacy Python for Everybody | freeCodeCamp.org
• Machine learning development is mainly done in a specific code editor called Jupyter Notebook. You can either install Jupyter Notebook on your PC or use online cloud-based Jupyter notebook platforms like Google Colab and Kaggle.
  • Jupyter Notebooks tutorial: (https://www.youtube.com/watch?v=3C9E2yPBw7s)
  • Google Colab (https://colab.research.google.com/): Free online Jupyter notebook environment with GPU support.
    • Colab tutorial : Google Colab Tutorial for Beginners | Get Started with Google Colab (youtube.com)
  • Kaggle (https://www.kaggle.com/): Platform for data science competitions, notebooks, and datasets. Also provides online Jupyter notebook
    • Kaggle tutorial: What is Kaggle & How to Use Kaggle? Kaggle Tutorial for Beginners - Full Walkthrough (youtube.com)

2. Data Manipulation and Visualization

• Learn NumPy for numerical computing and Pandas for data manipulation
  • Data Analysis with Python Certification | freeCodeCamp.org
  • tools_numpy.ipynb - Colab (google.com) Google colab notebook with numpy tutorial (Recommended)
  • tools_pandas.ipynb - Colab (google.com) Google colab notebook with pandas tutorial(Recommended)
• Explore data visualization with Matplotlib or Seaborn
  • Codebasics - Matplotlib tutorial - YouTube
  • tools_matplotlib.ipynb - Colab (google.com) Google colab notebook with matplotlib tutorial (Recommended)

** You can download or copy the given google colab notebooks in your own colab account.

3. Introduction to Machine Learning

**You can either follow one of the complete ML playlists or study the topics individually from various sources

Complete ML Playlists:

• Machine Learning | Coursera (Recommended)
• Codebasics - Machine Learning Tutorial Python | Machine Learning For Beginners - YouTube
• Krish Naik Hindi - Machine Learning Playlist - YouTube
• ageron/handson-ml3: A series of Jupyter notebooks that walk you through the fundamentals of Machine Learning and Deep Learning in Python using Scikit-Learn, Keras and TensorFlow 2. (github.com) A github repository containing jupyter notebook files with codes. You can download and open the files in colab. (Short and precise, good for revision or quick learning)
• CampusX - 100 Days of Machine Learning - YouTube

Topics:

** You can explore the given playlists to find specific topics or search for videos and articles by entering keywords related to your interests.

• Understand basic ML concepts

• Learn scikit-learn library

Basic Machine Learning Algorithms

• Linear Regression

• Logistic Regression

• k-Nearest Neighbors (k-NN)

• Projects:

  • House price prediction (Linear Regression)
    • https://www.youtube.com/watch?v=1Gx1y2boe_M
  • Diabetes prediction (Logistic Regression)
    • https://youtu.be/XCdwqybQWcw?si=dQhf56P5awzqQhkd
  • Iris flower classification (Logistics Regression)
    • https://youtu.be/n40hS9tQmcY?si=261m0ueVGfwdRZJC
  • Iris flower classification (k-NN)
    • https://youtu.be/E8Nrl9QLNq8?si=InC-kYuS0GYQDfnj

Advanced Machine Learning Algorithms

• Decision Trees and Random Forests

• Support Vector Machines (SVM)

• Naive Bayes

• K-Means Clustering

• Principal Component Analysis (PCA)

• Projects:

  • Customer churn prediction (Random Forest)
    • https://youtu.be/-8stRxINNeY?si=1qvuqyTHDcRkKG71
  • Spam email classification (SVM and Naive Bayes)
    • https://youtu.be/YncZ0WwxyzU?si=z4-CIuUihHA9TBVj
  • Customer segmentation (K-Means Clustering)
    • https://youtu.be/7J3mz9LMjqg?si=T8woaYRMLTAU-4Vz
  • Dimensionality reduction for visualization (PCA)
    • https://youtu.be/8klqIM9UvAc?si=ekKF_aKppzZx91yt

4. Deep Learning Fundamentals

• Understand neural networks basics

  • Nvidia Deep Learning Institute - Building A Brain in 10 Minutes (This notebook explores the biological and psychological inspirations to the world's first neural networks.)
  • Rajat Dandekar - Building Neural Networks from Scratch - YouTube (Recommended)
  • 3Blue1Brown - Neural networks - YouTube (You can watch this series for understanding the intuition behind Neural Networks)

• Deep Learning Theory

  • Krish Naik Hindi - Complete Deep Learning Playlist - YouTube
  • Stanford CS230: Deep Learning | Autumn 2018 (youtube.com) Lecture series by Prof. Andrew NG. Good for formal deep learning knowledge.

• Choose between TensorFlow or PyTorch

  • Tensorflow and PyTorch are two major frameworks used for deep learning development. You can start with Tensorflow as it has better community support and projects availability.

Complete Playlists

• Codebasics - Deep Learning With Tensorflow 2.0, Keras and Python - YouTube (Uses Tensorflow) (Recommended)
• Krish Naik - Live Deep Learning Sessions - YouTube (Detailed live stream tutorials)
• Learn PyTorch for deep learning in a day. Literally. (youtube.com) (Uses Pytorch)
• Dive into Deep Learning — Dive into Deep Learning 1.0.3 documentation (d2l.ai) (A complete book series on deep learning with tutorials and exercises. Good for book learners.)

5. Natural Language Processing

Complete Playlist

• Krish Naik - Live NLP Playlist - YouTube

• Projects:

• Sentiment analysis on movie reviews

• Text classification with BERT

• Question-answering system

6. Capstone Projects

• Choose 2-3 larger projects that combine multiple areas:

• Build a recommendation system

• Create an image captioning model

• Develop a language translation system

Continuous Learning

• Solve ML/DL problems and challenges

  • Sigmoid Academy (sigmoid-academy.netlify.app)
  • ML Code Challenges - Deep-ML ** platforms for solving coding challenges, similar to Leetcode

• Participate in Kaggle competitions

  • Kaggle Competitions

• Contribute to open-source ML projects

• Stay updated with ML/DL blogs and research papers