How Forward Propagation in Neural Networks works

[@mehmetakyuz1632]

This is awesome. Right level of detail balanced with simplification. Great work!

[@NM-vw6xq]

Misra, thank you so much for putting together these videos. They helped me out a lot. Seeing you walk through a simple example made a lot of things click for me vs. just seeing random notation in a textbook that I never fully understood.

[@misraturp]

Glad it was helpful!

[@jameelabduljalil25]

Thanks for the time and effort you put in these lessons and put them for free.. Really appreciate it..💐

[@misraturp]

You're very welcome Jameel!

[@HomeDesign_Austin]

great detail for new learners, great job

[@hshrestha2811]

You have presented things in a very simple and comprehensive manner. Thanks!

[@misraturp]

You're very welcome :)

[@zainulhassan8557]

Thank You so much Mam'm. Really appreciate your efforts. You explain all the things in an easy way.

[@behradio]

Simple and Understandable, Thank you 🙏

[@misraturp]

You're welcome :)

[@alexanderkamke3774]

Thanks a lot! Its very good to understand ...

[@viktorkovacs9680]

Thanks, I understand better now. But are the slides correct at 9:11 and 12:13?

[@bay-bicerdover]

Yep

[@prasad4685]

Thanks ma'am- very useful

[@SheepKev]

I'm a bit confused at 12:34, how does `(w3x1 + w4x2) + b2` turn into `(w1x1 + w2x2) + b1`?

[@joniendog6693]

Memang buat pemula seperti saya belajar ini terasa agak sulit ' biar cepet memahami saya terus mengulang ulang melihat videonya ' terima kasih ' semoga sehat selalu

[@deepthireddy1789]

Malay?

[@went-rogue]

so i first saw you Assemble Ai, loved your content, thank you so much

[@misraturp]

Glad you enjoyed!

[@dancode0775]

clear and precise thank you

[@misraturp]

Glad it was helpful!

[@hshrestha2811]

I think you need to have two columns in the W transpose matrix. They are now presented as a single column with w1w2 as a product form. For matrix multiplication, number of columns in W transpose matrix should be equal to the number of rows in X matrix (2).

[@bay-bicerdover]

Disagree

[@schorrer17]

Hey thanks for the great video! How would the formula be, if there would be a second hidden layer? assuming B would be the output of the second layer, W2^T would be the weight matrix of the second layer, b2 its bias and so on B=beta{W2^T*[alpha(W1^T*X+b1)]+b2} is this correct? thanks for any help

[@himanshupandey9902]

Great ....... keep continue.

[@misraturp]

:)

[@mohammedzia1015]

Please share the slides and course notes for Lesson 2, Module 1.

[@misraturp]

Thanks for the heads up. They're there now.

[@lakeguy65616]

RELU, if the value is less than zero, relu returns 0, if greater than zero, relu returns the value (unchanged).

[@bay-bicerdover]

Well?

[@lakeguy65616]

@@bay-bicerdover well what?

[@bkentffichter]

This is all fine, but I have no idea what all of this means as far as learning. I need to have some sort of practical something.