Sheesh. After many videos of backpropagation, this is by far the best explanation. I just wrote everything down on a whiteboard and everything made sense. Thank you, andrej. You are my GOAT

GPT watching these lectures eating popcorn

Can someone explain what's happening in 39 mint??

best explanation, thank you!

kzbin.info/www/bejne/imi8nIONpdx5epYsi=w5Zc3NyijbNQhc6L&t=3235 Attention at the time didn't deserve even a word.

Damn i would have stocked up on Nvidia stocks if i watched this in 2016 51:55

MINDBLOWING SIMPLICITY!!! Well explained!

Andrej is so good

This lecture feels like the lecturer is only reading the slide aloud. Explanations are unclear, the lecturer hastily skipped through many parts, without providing any clear explanation.

I can confirm from Future that Thaeano is dead, and Caffe is in comma

done

Garcia Elizabeth Clark Mark Wilson Brian

Hall Gary Allen Paul Young Brenda

@14:47

Much thanks!

if you're confused by the example at 19:50, it is very simple but can be confusing because in the slides the rule for the output size was defined as (N-F) / stride + 1. the lack of parenthesis in that definition causes the confusion as it really should be ((N-F) / stride) + 1 which will then give you the correct answer. wrong: (32 + 2 + 2 - 5) / 1+1 = 31/2? correct: ((32+2+2-5)/1) + 1 = (31/1)+1 = 32

Where the cats okay??

Why there aren't any captions with the video?

watching this 8 years later cause i cant understand what my professor was tryna explain (she doesnt know how to explain)

Always loved to hear Dr Fei Fei Li speaks even though this is nearly 10 years ago and she is still pretty hilarious.

Very good lecture, Andrej. Also, the student who interrupted Andrej by saying 'that was what I said' didn't have a good tone and annoyed me.

hey internet, i was here

andrej has the most unique way of telling things and generally they are more intuitive, what a maniac!

Just want to say a big Thank you, this course had helped me prepare the foundation for my PhD back in 2017. Literally watched it multiple times to internalise the concepts

and this is the 100th

this is the 99th comment

27:40

6:00

Amazing Course ! Thank you!

Started my journey and watching this course now. Couldn't attend Stanford University but still good to be in the class in 2024 : )

This is helping me quite a lot, thanks!!!

In 2015, the world didn't know about the problem of batch normalization

1:14:27 - The network never fully converges, but at some point you stopped caring. Because it has been 2 weeks and you are just tired. 😅

You are one of the best Andrej. You make learning so fun, with moments like 27:45 😄 Forever grateful.

This clears lot of doubts I had in my head. Thank you Andrej.

Can't thank Andrej, the cs231n team, Stanford enough. Thoroughly enjoy your lectures. Knowledge is one form of addiction and pleasure and thank you so much for providing it freely. I hope you all enjoy giving it as much as we enjoy receiving it.

1:09:10 it worked :)

does he speaks 1.5x by default 😛

At 39:00, what did he mean by jiggling the scores?

one of the best videos on RNN who explains the code perfectly from sratch.

This content here, is GOLD.

28:00

I love you Andrej ❤

This is a beautiful lecture - gave a very fundamental understanding of backward propagation and its concepts - I see backward propagation correlates to demultiplexing and forward prop corresponds to multiplexing where we are multiplexing the input .

haha. at 1.17.00 Justin says "it's cool but not sure why you would want to generate images"... 🙂

Seems to be a mistake in the "Computing Convolutions: Recap" slide - it says "FFT: Big speedups for small kernels" when it should be "big kernels"?

YOLO is king now hahaha.

when this video was recorded I could not even factor a quadrtic equation. Now I can watch this and follow the math w realetive ease. wow

Day -2 . 3 lectures and counting...

10