Lesson 12: Deep Learning Foundations to Stable Diffusion

[@giorda77]

Lesson 11 and 12 should be called "Broadcasting:the missing semester." Great content @Jeremy Howard. Thanks again

[@kiriyama0]

I can't express how amazing is that insight at 1:21:50 when Jeremy manipulates the dimensions in order to make the operation possible! I love this course so much!

[@bemjaxide6594]

Great work Jeremy, the simplicity of your explanations reflect great knowledge...just awesome

[@howardjeremyp]

I appreciate that!

[@mathumble6684]

1:31:00 the bs variable is never passe into the %timeit call at all, and a good practice also is to adopt using kwargs to avoid confusion

[@mathumble6684]

the original meaning of einstein notation in physics, only include the sum, it is pytorch that extended its meaning to have hanging un-summed elements.

[@ziolkovy]

Bug in 1:08:46?
while performing a loop in one_update, let's say i == 2, we update X[2] based on already updated X[0] and X[1], so we don't treat all points the same way. Sure, it works but doesn't sound like it is mathematically correct. To solve we should either calculate all weights for all points and then perform updating phase. Or create a temporary 1500,2 tensor as a returned result and in parent scope call X = one_update(X)

[@grekiki]

Agreed, also helps to determine why the GPU implementation doesn't quite give the same results as the CPU one.

[@krzysztofwos1856]

The meanshift algorithm that picks only the closest point would be equivalent to the mean field approximation often used in physics-based animation.

[@kiriyama0]

I am happy to know that Jeremy is a fellow lefty

[@SarisKiattithapanayong-hx3dw]

how can you make MAC book using CUDa?

[@mkamp]

Does somebody know which tool is Jeremy using to draw onto the screen?

[@dipamvasani8727]

oneNote

[@kiriyama0]

it's presentify (but only for mac)