/Machine-Learning-Resources

This repository contains the link to all the resources that I have used to develop my Python and ML knowledge.

Machine-Learning-Resources

This repository contains the link to all the resources that I have used to develop my Python and ML knowledge.

Python:

  1. YouTube: Complete Python tutorial in Hindi (2020):
    Url: https://www.youtube.com/playlist?list=PLwgFb6VsUj_lQTpQKDtLXKXElQychT_2j
    This is a very detailed course on python. I myself haven't completed it. I just use this to refer basic concept of python when I have doubts. Follow this course only and only if you are complete beginer.

  2. Udemy: Automate the Boring Stuff with Python Programming:
    URL: https://www.udemy.com/course/automate/ This is one of the most important course I have ever enrolled in. This knowledge from this course has helped me build 'cool' applications with real time use. If you are tired from writing fibonacci code with python and want to build something real and constructive like whatsapp automation or Web Scrapping Projects, enroll in this course.

Mathematics for Machine Learning:

  1. YouTube: MIT 18.06SC Linear Algebra, Fall 2011 By Gilbert Strang:
    Url: https://www.youtube.com/playlist?list=PL221E2BBF13BECF6C
    From dimension and basis to psuedoinverses this course covers more than what is required for machine learning.I credit this course a lot because having completed this course, the Andrew NG's course on ML felt very easy to me. This course is one of the unsaid pre-requisite course for Andrew NG's course.

  2. YouTube: Essence of linear algebra:
    Url: https://www.youtube.com/playlist?list=PLZHQObOWTQDPD3MizzM2xVFitgF8hE_ab
    The above mentioned Linear Algebra course by MIT is very detailed but at the same time very time consuming. A less lengthy alternative will be this course by Grant Sanderson. This video series is very graphic, visual, illustrartive and pictorial. Although I will warn the course can get very tough as the material is very dense.

  3. YouTube: Essence of Calculus:
    Url: https://www.youtube.com/playlist?list=PLZHQObOWTQDMsr9K-rj53DwVRMYO3t5Yr
    Again a brillant playlist by Grant Sanderson. A consice and visual illustration of calculus.

  4. edX: Probability - The Science of Uncertainty and Data:
    Url: https://www.edx.org/course/probability-the-science-of-uncertainty-and-data This is a MIT Micromaster Program so definitely it will be long and detailed. In my opinion if you have elementary level knowledge of probability, you can manage to go for ML courses. Enroll in this course only if you necessarily want to learn probability.

Machine Learning:

  1. Andrew NG
    a. Coursera: Machine Learning:
    Url: https://www.coursera.org/learn/machine-learning
    This course is the best course to understand the theory behind the ML algorithms. This course not only give idea about the algorithms of ML but also gives practical tips needed to implement ML model. Although the course is in octave(a programming language), I highly encourage everyone to also implement the code in python. This is a paid course but you can audit the course for free.

    Alternatively, you can find the same course on YouTube.
    b. YouTube: Machine Learning — Andrew Ng, Stanford University [FULL COURSE]:
    Url: https://www.youtube.com/playlist?list=PLLssT5z_DsK-h9vYZkQkYNWcItqhlRJLN
    I audit the course from YouTube and not from coursera because In feel that YouTube has easier interface than coursera. Also, the assignments are not present in YouTube so all I did was follow the video lectures and hence I could finish the course faster. However, when I completed the course, I went back to coursera and solved all the asssignments there.

    Another alternative, Ml course by Andrew NG in python.
    c. YouTube: Stanford CS229: Machine Learning | Autumn 2018:
    Url: https://www.youtube.com/playlist?list=PLoROMvodv4rMiGQp3WXShtMGgzqpfVfbU
    I haven't followed the course but I can vouch for the work of Andrew NG.

  2. Udemy: Machine Learning A-Z™: Hands-On Python & R In Data Science:
    Url: https://www.udemy.com/course/machinelearning/
    As much as I adore Andrew NG's course for it's theory, I cannot deny the fact that the course doesn't focus on implementation. Now, I haven't folllowed this course yet but many people have recommended me this course so it is worth a try.

Imp: Kaggle (https://www.kaggle.com/)
This is the place where we can find our dataset. Simply type in the search box and click on the results.