Soegyi/ml

Machine Learning & Deep Learning Roadmap

Machine Learning & Deep Learning Roadmap

Beginner

Understand basic concepts, learn Python, and be able to differenciate Machine Learning, Data Mining and Deep Learning

• Introductions
  • Interview with Tom Mitchell
  • A Gentle Guide to Machine Learning
  • What is the difference between Data Analytics, Data Analysis, Data Mining, Data Science, Machine Learning, and Big Data?
  • Data Mining Vs Artificial Intelligence Vs Machine Learning
  • A Visual Introduction to Machine Learning
• Get started with Python
  • Syntax, data types, strings, control flow, functions, classes, exceptions, networking, asynchronous task, function decorator, annotation, context manager, multiprocessing etc......
• Machine Learning Resources for Getting Started

  • Online Video Course

    • Build Intelligent Applications (Python)
    • Stanford Machine Learning (Octave)

  • Overview Papers

    • The Discipline of Machine Learning
    • A Few Useful Things to Know about Machine Learning

  • Beginner Machine Learning Books

    • Programming Collective Intelligence
    • Machine Learning for Hackers
    • Machine Learning, 2nd Edition
    • Data Mining: Practical Machine Learning Tools and Techniques, Third Edition
    • Practical Machine Learning

  • Complete at least the Online Video Course

  • Start a small project for creating a Python Web Crawler application and a RestFul Service to explore data stored

Novice

Get familiar with Python Machine Learning libraries

• Install and practice Python libraries
  • pip
  • asyncio
  • jupyter
  • scikit-learn
    • Quick Start Tutorial
    • User Guide
    • API Reference
    • Example Gallery
    • Papers
      • Scikit-learn: Machine Learning in Python
      • API design for machine learning software: experiences from the scikit-learn project
    • Books
      • Learning scikit-learn: Machine Learning in Python
      • Building Machine Learning Systems with Python - Second Edition
      • Machine Learning with scikit learn tutorial
  • scikit-learn is built upon the scipy (Scientific Python) includes:
    • numpy, base n-dimensional array package
      • Python Numpy Tutorial
    • scipy, fundamental library for scientific computing
    • pandas, data structures and analysis
      • Python For Data Science Cheat Sheet
    • sympy, symbolic mathematics
    • matplotlib, comprehensive 2D/3D plotting
    • ipython, enhanced interactive console
• Study scikit-learn, read documentation and summarize the capabilities of scikit-learn
• Linear Regression example in Python
• Linear Regression using scikit-learn
• Logistic Regression using scikit-learn
• Regression analysis using Python StatsModels package
• Using Logistic Regression in Python for Data Science
• Logistic Regression and Gradient Descent (Notebook)
• Regression analysis using Python StatsModels package
• k Nearest Neighbours in Python
• An Introduction to Supervised Learning via Scikit Learn
• An Introduction to Unsupervised Learning via Scikit Learn
• Start a project to implement a simpler algorithm like a perceptron, k-nearest neighbour or linear regression. Write little programs to demystify methods and learn all the micro-decisions required to make it work

Intemediate

Learn Neural Networks and understand Deep Learning

• Online Video Courses

  • Neural Networks for Machine Learning

• Books

  • Deep Learning
  • Deep Learning Tutorial
  • Neural Networks and Deep Learning

• Papers

  • ImageNet Classification with Deep Convolutional Neural Networks
  • Deep Learning in a Nutshell: Core Concepts

• Study one of the Machine Learning Dataset from data.gov

  • Clearly describe the problem that the dataset represents
  • Summarize the data using descriptive statistics
  • Describe the structures you observe in the data and hypothesize about the relationships in the data.
  • Spot test a handful of popular machine learning algorithms on the dataset
  • Tune well-performing algorithms and discover the algorithm and algorithm configuration that performs well on the problem

• Design small experiments using the Datasets for studying Linear Regression, or Logistic Regression, then answer a specific question and report results

• Try to port an open source algorithm code from one language to another

Advanced

Get to know the Python Frameworks for Deep Learning, and focus on TensorFlow

• Study Neural Networks in Python
  • Implementing a Neural Network from scratch in Python
  • A Neural Network in 11 lines of Python
  • Get familiar with the Neural Networks libraries
    • Caffe, a deep learning framework made with expression, speed, and modularity in mind
    • Theano, CPU/GPU symbolic expression compiler in python
    • TensorFlow, an open source software library for numerical computation using data flow graphs
    • Lasagne, a lightweight library to build and train neural networks in Theano
    • Keras...... check the links below:
  • Deep Learning Software Links
  • Check rest of Deep Learning Libraries by Language
• Deep Learning With Python
  • Study TensorFlow
  • Study Keras, a high-level neural networks library, which allows for easy and fast prototyping (through total modularity, minimalism, and extensibility)
• Books
  • Hands-On Machine Learning with Scikit-Learn and TensorFlow
  • First Contact with TensorFlow
  • Getting started with TensorFlow
• Videos
  • Videos from TensorFlow Silicon Valley Meetup, Jan 19, 2016
  • Videos from TensorFlow Seattle Meetup, Jan 21, 2016
• Code Samples
  • Learning TensorFlow
• TensorFlow knowledge points
  • Graph, Session, Variable, Fetch, Feed, TensorBoard, Playground, MNIST Practice, APIs
  • Linear Regression, Logistic Regression Modeling and Training
  • Gradients and the back propagation algorithm, Activation Functions
  • CNN, RNN and LSTM, DNN
  • Unsupervised Learning, Restricted Boltzmann Machine and Collaborative Filtering with RBM
  • Auto-encoders, Deep Belief Network, GPU programming and serving

Version 0.3, by Michael