This my solutions to all the assignments of the coursera course tought by Andrew Ng.

in this repository, I've implemented solutions in both 2 languages octave and python and using jupyter notebook for better project structure and better documentation.

In the course materials and octave, 1D vector always refers to 1D column vector, but in many python libraries they seems to favor 1D row vector.

2 possible solutions

1. use remote connection
2. could be virtual env python path problem, the octave kernel config by default doesn't specify which python to use, make sure you are in the virtual env environment or just specify python version in the config.

It means shell bracked paste mode is on. it not only generate special code to the screen but also cause performance drop.

copy the jupyter notebook config file to jupyter's config folder to solve this problem.