Hello! I am Joe from Taiwan.
I’m picking up Andrew Ng’s ML class on Coursera, which I first took back in 2015, where we used Octave (although I think I used MATLAB).
I’ve been studying math all these years, but now looking for a job related to AI.
1 Like
The Machine Learning Specialization is an update to the original course, using more modern tools and improved lectures.
1 Like
It seems like so!
On one hand I’m glad that I’ve learned Python syntax recently so that I may catch up, while on the other hand I have little idea how each lab assignment is built—apparently from a package of their own and some public ones, but I don’t know how exactly each of them works, which makes me feel weird.
Some assignments have you write your own tools (like the old ML course did), but they graduate rapidly to using sklearn or TensorFlow. Similar to how the ML course switched to using pre-made optimizers like fminunc() and fmincg() around Week 3 or 4.
Part of the goal is to spend more time creating models using modern tools. Creating the model is where the interest lies, not in writing code that does optimization.
1 Like
Hello Joe, welcome back to MLS!
Cousera’s jupyter interface allows you to check out those custom script files used in the lab (File Menu > Open), so you can actually open those script files and examine the code.
Cheers,
Raymond
1 Like
Wow I didn’t know that! Now I can see the hidden packages! Thanks for sharing!!
1 Like
You are welcome, Joe!