0 like 0 dislike
1,301 views
in ML/AI by Expert (15,690 points) | 1,301 views

3 Answers

0 like 0 dislike
0 like 0 dislike

FREE RESOURCES : 

1)https://www.codewithharry.com/videos/python-tutorials-for-absolute-beginners-99

2)https://www.coursera.org/learn/machine-learning

3)https://www.mygreatlearning.com/blog/machine-learning-tutorial/

4)https://www.youtube.com/watch?v=GwIo3gDZCVQ

 

ROADMAP : 

Keep in mind the following points before you dive into Machine Learning(ML) with Python :

  1. Like any other programming languages, you need to learn Python too, before you get started with ML with python.
  2. Learning Python WILL take time, do not think that you can be an expert programmer in a day!
  3. There is NO Machine Learning without Linear Algebra, Statistics and Probability theory.
  4. The best way to learn to code is to do small tasks.

Now, we start the roadmap to ML,

  1. Learn basic Python:

    Go to Codeacademy and start learning, if you spend around 5 hours a day, you will be done with basic Python in around 3–4 days.

    I would also like to suggest getting the Learn python the Hard way (avaibale for free online here)book. This book is amazing.
  2. Learn basic data structures using Python:

    For this go to Geekforgeeks-Data structures, open each data structure and go to the Python tab. See how they implement it, learn from them.

    DO NOT JUST MINDLESSLY LOOK AT THE CODE, write the code and learn from it.

    This process should take around 5–7 days, if you work for around 5 hours a day.
  3. Learn some cool Python packages:

    Some of the popular ML packages (basic) used in Python are as follows:

    * Numpy.
    * Pandas.
    * Matplotlib.
    * Scikit.

    Learn each of them; Read from the official documentation of the packages. See the various implementations, look for the WHYs and the WHY NOTs.

    Should take about a week.
  4. Do basic Linear Algebra, Statistics and Probability with Python:

    Use the above packages and do some basic math with it.

    Follow the following links:
    Linear Algebra with numpy.
    Statistics and Plotting with matplotlib.
    Python, Numpy and Probability.
    10 minutes into Pandas.

    Please find more on your own and learn, send an edit request to the list, and if I find the change substantial, I will approve it.

    This step must take around 3 to 5 days.
  5. Start with Machine Learning (Now this is the scary part :p):

    We start learning ML now, for that we need the required Theoretical Background, follow the CS156 from Caltech by Professor Yaser Abu-Mostafa. The youtube link to his Machine learning lectures are here : CS156 - Machine Learning.

    A week must suffice.
  6. Start implementing Machine Learning algorithms:

    Hands down, the best place to learn this is scikit-learn.

    Read this blog to expand your horizon : Machine Learning with JSAT

    Follow Kdnuggets and AnalyticsVidhya.

    Do small problems from Kaggle and use Google extensively.

    A week will suffice.
  7. Start getting smarter — Read a lot.

    Get started with awesome technologies and packages like :

    * H2O.
    * PyLearn.
    * PyBrain.
    PyPI packages with “Machine Learning” in their description (a whole big list)
by Expert (15,690 points)
0 like 0 dislike

Video Courses

There are an overwhelming number of video courses and tutorials available online now — many of them free. There are some good paid options too, but for this article, I’m focusing exclusively on free content. There are considerably more college courses where the professor has made the course materials available online, but there are no videos. Those can be more challenging to follow along and you probably don’t need them. The following courses would keep you busy for months:

YouTube

Below I include links to YouTube channels or users that have regular content that is AI or machine learning-related. I’ve ordered by subscriber/view count to give a sense of their popularity.

Blogs

Given the popularity of AI and machine learning, I’m surprised there aren’t more consistent bloggers. Given the complexity of the material, it takes quite a bit of effort to put together meaningful content. Also, there are other outlets like Quora that give options to experts that want to give back but don’t have the time to create longer form content.

NLP

Math

Tutorials

I created a separate comprehensive post covering all the good tutorial content I’ve found:

Cheatsheets

Similar to tutorials, I created a separate article with a variety of good cheat sheets:

 

by

Get best answers to any doubt/query/question related to programming , jobs, gate, internships and tech-companies. Feel free to ask a question and you will receive the best advice/suggestion related to anything you ask about software-engineering , development and programming problems .