How to Represent a Neural Network with Matrices

Subscribe for more content ► bit.ly/2Lf16p1
This video describes the basics of what a fully connected neural network is and how to represent it with matrices.
An implementation in Swift coming soon.

[@sarahlewicki7268]

I have been searching for someone to explain the relationship between matrices and neural networks and this one is by far the easiest to understand. Thank you so much!!!!

[@StevenDeville-tm9vv]

This guy knows his stuff. So much I gained watching this video.

[@jarhatz]

Great visualization man! This was literally just what I was looking for. Good stuff.

[@bishwarup1429]

1 year into ML still got a lot of things sorted from this video. Thank you chief. :)

[@BrianOSheaPlus]

Thanks, this is a really helpful visualization. For anyone who is taking Andrew Ng's deep learning courses on Coursera, keep in mind that he represents the matrices transposed relative to the way they are shown here, but otherwise are the same.

[@Jeabinoux]

This is a really good explanation. The only thing missing would be the bias applied to each neuron and how it participates in the calculation. Still great video!!

[@abdelghanyaref8230]

you are really super awesome dude ... i have been searching for two days and ur video was the one to solve all the mystery ... thank you so much

[@BioHazarddasdadfasfsad]

Sir this is brilliant! Searched "imagining neural network as matrix" and got here

[@Ab-cj6gl]

damn i wish the owner of this channel continued explaining stuff, top notch explanation thanks

[@emrehavan7456]

Awesome! I always felt bad because my strongest language is Swift (due to starting programming with iOS platform) but I was not able to get involved with CS stuff much (especially neural networks, gpgpu, parallel programming) other than implementing some data structures. But now this introduction to Metal & Neural Networks combined with Swift is great! You are a savior!

[@oliverford5367]

Swift isn't too different from Python, which is the standard language for ML front-ends. The guts though are in C++ and GPU native code which is harder. Python though has pretty good interfaces for all that.

[@liuqing1995]

very straightforward

[@jordancoinjackson7844]

Great channel just subbed looking forward to the full swift implementation 🔥

[@BananthahallyVijay]

Just a brilliant separation of wheat from chaff. 5 stars!

[@bjornnorenjobb]

This was exactly what I was looking for, big thanks!

[@amjadiqbal478]

ذبردست ❤❤

[@fairsnaillz-1293]

My brain grew 3 sizes that day.

[@mueez.mp4]

Very intuitive explanation.

[@Muuip]

Great concise presentation!
Much appreciated! 👍

[@shaukat9823]

this is great, but you did not mention the biases it it, could you please make another video about it ?

[@ooffoo5130]

THANK YOU

[@ramavenkatesh4215]

amazing video

[@debrahul2001a]

I have question on this. How can I send you my questions

[@WPFES]

Thank you! What happens when the input Data is 4 Dimensional eg. Image Classification with 3 Channels per Picture. How does the Matrix Multiplikation work?

[@gandalf9526]

Thanks a lot, subbed :)

[@ramkashyap2528]

I have been looking for something to clear my doubts, Thanks a ton.
it even solved things that I could stumble upon.
Subscribing Channel. would be looking for more of these videos.

[@vkray]

Waiting for your next video.

[@vaibhavbhasin1221]

you didn't tell about adding bias.

[@thomasdartez183]

When are you gonna upload more videos we are waiting!

[@alyology0]

why did u stop making vids? :(

[@HardcoreMontages]

This video should have negative dislikes