0

I want to get into machine learning. I've been in information security for the last 10 years so i have an IT background.

Where is the best place to start: Can anyone recommend a good book? And also a platform i can use to practice (preferably free) Also if there are any online courses someone could recommend that would be great.

I looked into AWS's offering of machine learning but that is not included in the free tier.

Any help/advice would be much appreciated.

Thank You.

Architect
  • 121
  • 2

2 Answers2

1

Courses: Andrew Ng Machine Learning Course from Coursera is what introduced me to machine learning, and I cannot think of a better way to do it. The course focuses on the theory rather than practice, but it offers solid theoretical foundations. fast.ai Machine Learning course is more programming oriented, I haven't taken it but it looks good.

Books: Hands on machine learning with Scikit-learn and Tensorflow by Aurélien Géron has a decent amount of theory explained in a very simple way, and it complements it with lots of code examples. If you want to dig deeper in statistics and theory, The Elements of Statistical Learning might be a good resource.

Practice: Kaggle is for sure the best place to practice and learn. After reading something or doing some course I advise to download the Titanic dataset and try to submit predictions.

David Masip
  • 6,051
  • 2
  • 24
  • 61
  • fast.ai is not at all for complete beginners as they will get lost in the forums.., the course is awesome as I have watched it online completely, It starts from RF all the way to NN, linear regression etc.. – Aditya May 31 '18 at 09:05
  • Yes, I guess you are right. The DL ones in fast.ai are quite hardcore. – David Masip May 31 '18 at 09:08
1
  • Online Course: Andrew Ng, Machine Learning Course from Coursera.
  • Book: Tom Mitchell, Machine Learning, McGraw-Hill, 1997.
Christos Karatsalos
  • 830
  • 4
  • 12