How to code neural networks without math*

When I code neural networks, I think about them write them in layers, often not thinking about the math that runs behind the scenes.
Yes, neural networks are completely based on math, though it's not 100% required to know everything happening behind the scenes to code and use neural networks.
So if you're thinking about getting into machine learning but being held back by the math, try writing algorithms in code first and learning math as you go, just like learning another programming language (math is Nature's programming language).
Original video source - • Is 13 too young to sta...
Meme source - marksaroufim.substack.com/p/w...
Links:
Learn ML (beginner-friendly courses I teach) - www.mrdbourke.com/ml-courses/
ML courses/books I recommend - www.mrdbourke.com/ml-resources/
Read my novel Charlie Walks - www.charliewalks.com
Connect elsewhere:
Web - dbourke.link/web
Twitter - / mrdbourke
Twitch - / mrdbourke
ArXiv channel (past streams) - dbourke.link/archive-channel
Get email updates on my work - dbourke.link/newsletter

[@mrdbourke]

*yes, neural networks are math functions but you don't need know the math inside out to start coding them. I've spoken to many people who've been scared of getting into machine learning because of the "math". School leaves a bad taste for learning math in many people's mouths. If that's you, and machine learning interests you, you shouldn't let it stop you from getting into it. If companies can use neural networks to provide you services/manipulate what you see on the internet/drive your cars, it's not a bad idea that you look inside these algorithms and realize that behind the scenes often they're a stack of a few lines of Python code with large amounts of data being passed through them. Like how driving a car doesn't require full knowledge of how a combustion/electric engine works, getting started with machine learning doesn't require a full understanding of how the math works behind the scenes (many of the smartest machine learning researchers will tell you they don't quite know how certain algorithms do what they do). When I think about driving a car, I move the steering wheel and press the pedals, not how fuel explodes in the engine and moves pistons up and down. When I write neural network code, I stack layers together and think about what inputs and outputs I desire, not every function running under the hood. I stole the meme in this video from Mark Saroufim's article Working Class Deep Learner - marksaroufim.substack.com/p/working-class-deep-learner For full context, this video is a clip from a livestream Q&A I did for 97,777 subscribers - kzbin.info/www/bejne/haXdn35tfpZkqZo

[@masternobody1896]

Nope it's all DNA not code

[@TundeBot]

Nope stop being a wuss and learn the math lmao

[@sravan8643]

congrats on 100k. Following you since you had like 800 subs!!!! WHAT YOU HAVE ACHIEVED IS REALLY GREAT

[@EvgenSuit]

Congrats for 100k subs, I wish you luck with further development of the channel !

[@mrdbourke]

Thank you Kip! Plenty more to come :)

[@malikhamza9286]

Hi, Congratulation for 100K Subs in Advanced. Happy for you!!!

[@mrdbourke]

Thank you Malik!

[@Frost_Byte_Tech]

I bought your course on Udemy and have followed through, am about to do the "last" part series and forecasting but I already feel like a beast when it comes to Neural Networks 💪💪 Thanks so much!!!

[@mrdbourke]

Outstanding effort Kokesto! Keep learning legend!

[@donatellosnizzo100]

Linear Combinations ?

[@ashikshafi1765]

Haha much needed 😂

[@lakshanjayawaradana2024]

99.9k yayyy! maybe tomorrow is the day with 100k WE WANT A PARTY DANIELLLL!!!!

[@mrdbourke]

There will be a party!! Don’t you worry about that, I’ve got something planned :)

[@lakshanjayawaradana2024]

@@mrdbourke YES!! CAN'T WAIT

[@erenyeager4452]

It's not Matrix Multiplications. It is 0's & 1's

[@AnushkaChathuranga-cw7tc]

How do you code neural networks with 0's and 1's ?