Here's what your first 30 days of Machine Learning should look like.

(I wish I had this before)
πŸ§΅πŸ‘‡
What you are going to look at now is a curriculum I wish I had followed.

It took me 2 years to get into machine learning because I was never consistent, however you don't have to make the same mistake.

You'll learn a lot in a very short amount of time through this thread.
Before we begin with the curriculum, make sure you throw these myths out of your mind.

> Machine learning does not require creativity

> Machine learning is difficult

> You need to know a lot of math to get into machine learning

> You need an expensive PC for machine learning
Also keep these points in mind.

> Machine learning is a vast field, the point of this thread is to help you get started, you'll have to keep that momentum going on afterwards too!

> Consistency is everything, it will decide if you use this curriculum well or not!
> Just because we will not being using any math during this time doesn't mean its not important, it is VERY important but it can come much later on in our journey.

> All courses and other material in this thread are 100% free.
With all that cleared, let's take a look at the curriculum you'll be following for the next one month.

Our goal will be to have a solid understanding of neural nets work without the complex math and we'll also learn how write quite a bit of code in this process.
We will also try to complete a simple kaggle challenge if possible.

You can read more about Kaggle here:πŸ‘‡

Step 1

Your first 10 days will be spent learning Python from this tutorial

πŸ‘‰www.​youtube.​com/watch?v=rfscVS0vtbw

Believe me, this is enough to get started with Machine learning, you can learn more complex python concepts when you need them.
30-45 minutes a day and you'll finish this course in about a week and a half or so, take longer if you need to, everyone has a different learning rate.
Here's a pro tip, incase the setup for python is too hard, you can use repl.​it, an online editor where you can write Python code without any setup. make sure you are using Python3 (3.7/3.8 is recommended) and not Python 2.
Step 2

We'll now dive in into how neural networks work!

This series by 3blue1brown has a simple, intuitive and a visual approach to how neural networks works without ANY complex math at all

πŸ‘‰www.​youtube.​com/watch?v=aircAruvnKk&list=PLZHQObOWTQDNU6R1_67000Dx_ZCJB-3pi
It won't take you longer than 3-4 days to complete this series.

I must also emphasise that along this journey you'll have to enjoy everything otherwise you will not be able to learn things as quickly as you could.
Step 3 : We're now about 17 days into our machine learning journey, here's the course you have to take next.

πŸ‘‡
Machine learning foundations by google developers, I can't recommend this course enough, it helped get started with machine learning in a way no other course could.

πŸ‘‰www.​youtube.​com/watch?v=_Z9TRANg4c0&list=PLOU2XLYxmsII9mzQ-Xxug4l2o04JBrkLV
These set of 12 videos in 12 days will take your machine learning skills from zero to 1, it really is a great course to take.

Did I mention you'll also learn how to use Google Colab and Jupyter notebooks in these coursesπŸ˜‰
We have one day remaining!

Our goal today is try and complete a practical project on kaggle, here's the competition you want to compete in as a beginner : Digit Recognizer challenge.
You've already trained a MNIST classifier in the google developers course, you'll have to apply that in the practical world using this challenge.

πŸ‘‰ www.​kaggle.​com/c/digit-recognizer
Tip : Take help from submissions from other people incase you are stuck. Also do not forget to google things when needed!
Our one month of machine learning has ended and we've learnt quite a bit and that is awesome!

I wish you all the best in all your future endeavours. πŸ”₯

β€’ β€’ β€’

Missing some Tweet in this thread? You can try to force a refresh
γ€€

Keep Current with Pratham Prasoon πŸš€

Pratham Prasoon πŸš€ Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @PrasoonPratham

20 Nov
There are a lot of myths surrounding machine learning.
When I say a lot, I mean it.

Here are 3 of the most common ones.

πŸ§΅πŸ‘‡
> π˜”π˜’π˜€π˜©π˜ͺ𝘯𝘦 𝘭𝘦𝘒𝘳𝘯π˜ͺ𝘯𝘨 π˜₯𝘰𝘦𝘴 𝘯𝘰𝘡 𝘳𝘦𝘲𝘢π˜ͺ𝘳𝘦 𝘀𝘳𝘦𝘒𝘡π˜ͺ𝘷π˜ͺ𝘡𝘺

This is just straight out wrong.
You have to be creative in finding solutions to obscure problems which you may encounter in machine learning ( and you can build this creativity! )
> π˜”π˜’π˜€π˜©π˜ͺ𝘯𝘦 𝘭𝘦𝘒𝘳𝘯π˜ͺ𝘯𝘨 π˜ͺ𝘴 π˜₯π˜ͺ𝘧𝘧π˜ͺ𝘀𝘢𝘭𝘡

No it isn't. You have thousands of resources online to help you out, not to mention you don't even need crazy hardware to train your machine learning models, all you need is a computer, an internet connection and passion.
Read 5 tweets
20 Nov
I recently got the opportunity of working at a start-up with a very good pay, even for a college graduate.

I declined the offer.

Here's why.
πŸ§΅πŸ‘‡
Here's a bit of the backstory, I was approached was the co-founder of this start-up through a twitter DM. The start-up is more than 1 year old at this point and was generating revenues of $250K + a year and was profitable! (at least what I was told in the call)
The start up is also based in India and I was offered a remote dev job as a machine learning engineer. This seems like a very good offer but ended up declining it.

Why?πŸ‘‡
Read 7 tweets
18 Nov
Some of the best free machine learning resources I found out over the last 2 years.

πŸ‘‡
Machine Learning Glossary by Google developers
> Has pretty much every definition and explanation for commonly used machine learning terms.

πŸ”—//developers.google.com/machine-learning/glossary
Towards DataScience
> A brilliant resource for reading articles about machine learning and data science

πŸ”—//towardsdatascience.com/
Read 5 tweets
18 Nov
Machine learning Basics πŸ€–

Pandas is one of the most important frameworks you'll be working with if you plan on getting into machine learning.

What is it and where is it used? πŸ€”

πŸ‘‡
πŸ—ƒοΈ In machine learning, you'll often come across data stored in data stored in excel sheets ( or similar kinds of data ).

In order to interact with that data you'll have to use "pandas". 🐼
πŸ“‚ Pandas allows you to read the data stored in these files without changing the original file itself.
Read 4 tweets
17 Nov
Programming is an essential part of Machine learning.

This is why I have decided to start Python quizzes inspired from @ravinwashere's quizzes to boost your problem solving skills.

Here's the first one, write python code that gives the output based on the input.
The first 5 participants who participate (doesn't matter if its wrong or right) will get a shout out πŸ˜‰
This was my attempt
(I know its horrible)πŸ˜…
Read 5 tweets
17 Nov
I made a lot of mistakes during my machine learning journey.

Being a web developer, here's why it took me 2 years to get started machine learning when I could've done it in less than a month.

(and you can too!)
πŸ§΅πŸ‘‡
My first attempt of getting into machine learning was in early 2018. I saw a few videos on YouTube but never really understood anything about it.
Terms like "Support Vector machines", "generative adversarial networks" always scared me.

Moreover there was nobody to guide me in the right direction.
Read 21 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!