demonstrate how to do regression from scratch in python with numpy for PGH Data Science Meetup.
- conda
the following step isn't necessary, but is useful for managing / switching kernels when launching jupyter notebooks from the default environment.
conda install nb_conda -q --yes
The following will set up the environment containing the requirements for the notebook.
conda create --name=pgh_data_science python=3.5 --file=requirements.txt --yes
which will create the environment for regression from scratch. to switch to that environment use
activate pgh_data_science on windows or source activate pgh_data_science on linux