Hacker News Reader

I'm guessing the author hasn't a maths education beyond high school.

I have a bachelors degree in economics, so some university math with calculus and linear algebra. The reason I found it hard was because I had to learn Octave while also understanding the algorithms and general ml concepts.

That being said, I think it's a great course, and would highly recommend it to anyone who thinks they're up for it. But there are other courses that provides an easier gateway into ML.

Like which one?

Udacity's Intro to Machine Learning

For another option, similar to the Udacity course, I recommend this: http://www.dataschool.io/machine-learning-with-scikit-learn/

> The reason I found it hard was because I had to learn Octave while also understanding the algorithms and general ml concepts.

I had the same experience, even after using Matlab some in grad school. I wish I had seen this before trying Ng's course (all of the exercises in Python):

http://www.johnwittenauer.net/machine-learning-exercises-in-...

So much quicker / easier to follow, at least for me.

scipy is almost exactly the same than matlab/octave

You shouldn't need scipy for this. Perhaps you meant 'numpy'?

Reply on news.ycombinator.com