IE MBD OCT 2018 | May 2019
Martin Hofbauer

Python Individual is a project which bases on a Maschine Learning project on Capital Bike Sharing in Washington DC.
The task was to reproduce the initial project using Dask in order to enhance the knowledge on parallel computing.

Most of the files are saved in a GitHub repository. The dataset can be accessed at Kaggle.
For ease of use, the dataset is also loaded into a seperate Gist so that when running the notebook, the download of the dataset runs automatically.

In order to be able to run the Jupyter Notebook, some preliminary installations have to be performed.

To install the necessary packages with Terminal (macOS):

• Run pip install dask
• Run pip install dask_ml
• Run pip install distributed
• Check if the package has been installed using conda list

To install the necessary packages with command prompt (windows):
Supposing that Python3 is installed and running.

• Run python -m pip install dask
• Run python -m pip install dask_ml
• Run python -m pip install distributed
• Check if the package has been installed using python -m conda list

To install the necessary packages with Anaconda Prompt:

• Run pip install dask
• Run pip install dask_ml
• Run pip install distributed
• Check if the package has been installed using conda list