Setting_up_Python.md

Setting up your python environment

The script has been tested on Python 2.7.14 and 3.6.3 and 3.6.4. The notebook has only been tested in 3.6 so prefer Python 3 if you don't have an environment yet.

Using pip

To ensure you have the necessary dependencies you can type this pip install command (whether in 2.7 or 3.6):

pip install requests matplotlib numpy pandas seaborn jupyter lxml beautifulsoup4 df2gspread docopt tqdm

On Windows, the preferred alternative is to use conda, but if you insist on using regular python with pip, you may have to manually install numpy+mkl.
You can do this by downloading the correct numpy+mkl wheel file for your platform from Gohlke's PythonLibs and then running:

pip install numpy-1.14.0+mkl-cp37-cp37m-win_amd64.whl

Using conda (Preferred for Windows)

Conda is available on all platforms, but on Windows I strongly advise you use it in order to avoid missing binaries etc.

On Windows, unless you have already installed python by yourself, the easiest is to use conda as a package manager (conda is both a package manager and an environment manager), which will take care of the required binaries etc.

You can choose between Anaconda or Miniconda. You can read more in conda's user guide, section Downloading conda, Anaconda comes prepackaged with hundreds of scientific python packages and takes about 300MB of disk space, while miniconda is a barebone option.

Whether you choose Anaconda or Miniconda, select the Python 3.6 version, and install that. I suggest taking a quick look at conda's Getting Started.

For setting up the environment, you have two choices:

• using the environment.yml file to create a dedicated environment that has only the required dependencies, or,
• just install the needed dependencies in the environment of your choice.

Using the environment.yml file

There is a file OpenStudio-resources/environment.yml that you can use to create an environment ready to be used. Go to the root of this project and type:

conda env create -f environment.yml

This will create an environment named openstudio-resources that you will need to activate every time:

conda activate openstudio-resources

You should see the prompt change, and you can type conda env list and there should be a star * next to it.

If you use git bash, you can place the following in your ~/.bashrc file to automatically activate it when you enter the OpenStudio-resources folder:

# Adapted from conda-auto-env 
# Automatically activates a conda environment when
# entering a folder with an environment.yml file.
#
# If the environment doesn't exist, creates it and
# activates it for you.
#
function conda_auto_env() {
  if [ -e "environment.yml" ]; then
    ENV=$(head -n 1 environment.yml | cut -f2 -d ' ')
    # Check if you are already in the environment
    if [[ $PATH != *$ENV* ]]; then
      # Check if the environment exists
      source activate $ENV
      if [ $? -eq 0 ]; then
        :
      else
        # Create the environment and activate
        echo "Conda env '$ENV' doesn't exist."
        echo -e -n "Do you want to create it with conda? [y/N] "
        read -n 1 -r
        echo    # (optional) move to a new line
        # Default is No
        if [[ $REPLY =~ ^[Yy]$ ]]; then
          conda env create -f environment.yml
          source activate $ENV
        fi
      fi
    fi
  fi
}
export PROMPT_COMMAND=conda_auto_env

Manually setting up your environment

After you are in the environment of your choice, you can run these install commands (if you have installed Anaconda, you already have most of them but it won't hurt).

conda install requests matplotlib numpy pandas seaborn jupyter lxml beautifulsoup4 docopt tqdm
conda install -c conda-forge df2gspread