Intro to ML or directly jump on DL?

(Giraldi Antoine) #1

Hello Fast.ai world, I am completely new to this Machine Learning world… So I have one question as this course is still in its 2018 version compared to Practical Deep Learning for coders which is in version 3.

What is your advice if I want to start ML from scratch knowing that I am an EE engineer. I have experience in control systems. Started with the first four videos of intro to ML… should I switch to practical deep learning for coders?

Thanks for advice Fast.ai community :slight_smile:

0 Likes

(Mahdi ) #2

Hello
i think you should keep watching ML for coders
in latest parts of that Jeremy was doing feature engineering for rossman competition
and he used it’s data for i think lesson 7 in deep learning for coders

ML course has some definition which is basic and necessary for doing any project in DL
if you have enough time , watch ML course first
you will benefit from that

i have not watched V3 , i have no Idea about that

but i think there is some issues in DL2 (2018) which Jeremy has not taught those in V3

0 Likes

(Giraldi Antoine) #3

Thanks Mahdi for your advice! I’ll finish the course. Just wondering if this is not better to switch to DL lecture after lesson 5 which is the end of RF part.

Best

0 Likes

(Mahdi ) #4

I have also down that at first
but if you do that you would miss lost of points about pre processing and feature importance which is necessary for every DL or ML projects
which should be conducted by some fast model like Random Forset

0 Likes

(Mahdi ) #5

feature engineering i mean

0 Likes

(Giraldi Antoine) #6

Ok thanks. I will continue this course to the end so

0 Likes