Thread by Harpreet Sahota 🥑

twitter.com
1 Recommender
1 Mention
Recommend
Post
Save
Complete
Collect
Thread
Here's how I would study deep learning if I had to do it all over again.

#deeplearning #machinelearning
👇🏽 🧵
1) Skip the math

I’d ignore the math when first starting out.

Looking at equations will demotivate you.

Instead, look for applications of deep learning.

Clone the
@ultralytics
Yolov5 repo and run the usage on the command line.

See the magic happen.

Get inspired.
2) Go through @AndrewGlassner’s DL crash course.

It’s 3.5 hours long but will give you an intuition for how it all works under the hood.

Great return on time investment.

www.youtube.com/watch?v=r0Ogt-q956I&feature=youtu.be
3) Go through @mrdbourke's free zero to mastery PyTorch course.

Now that you’ve got a bit of intuition under the hood, start coding.

It’s free and completely self paced.

You’ll learn how to code in @PyTorch - the best for DL imo - in about a week.

www.learnpytorch.io/
4) Now go and get into some math

My friend @JonKrohnLearns has an amazing course on YouTube that will teach you the basics of DL math.

It’s well worth the 2.5 hour investment.

www.youtube.com/watch?app=desktop&v=wBgW3ZtlPT8
5) Grok Backprop

You need to understand backprop.

And for that @AndrewGlassner has an amazing resource. No math. Just visuals.

It’s free and will take a couple of hours to read:

dlbasics.com/resources/Backpropagation/Chapter18PDF.pdf
5.1) Go through @karpathy's YouTube video! Learn how backprop works by coding it from scratch.

www.youtube.com/watch?v=VMj-3S1tku0&t=101s
5) Get an understanding of the foundational works.

For that I recommend @ykilcher’s YouTube series on classic papers.

www.youtube.com/playlist?app=desktop&list=PL1v8zpldgH3qQB5Pz6ZSTTDLu0BjAJYNf
6) Join a community of learners

You want to be around a range of learners, from experts to beginners.

For that I’d recommend joining a community.

Deep Learning Daily is small at the moment, but that’s good cuz you’ll get help if you ask questions.

www.deeplearningdaily.community/
7) Do projects

Now it’s time to take off the training wheels and do projects.

As many as you can.

Share your work, get feedback, improve.
That’s it. That’s how I would learn DL if I had to all over again.

If you want to get even deeper in the theory, then I’d too Stanford’s CS231 in the mix. www.youtube.com/watch?app=desktop&v=vT1JzLTH4G4

I'd also read this blog, A Recipe for Training Neural Networks : karpathy.github.io/2019/04/25/recipe/
What else would you add to this?

1. Follow me @DataScienceHarp for more of these
2. RT the tweet below & share w/ your friends
3. Also follow: @nevrekaraishwa2 @SanthoshKumarS_ @GiftOjeabulu_ @Sumanth_077 @Saboo_Shubham_ @_jaydeepkarale @Pauline_Cx

Mentions
See All