Crime and Settlement
[Case Study] This project offers an interactive map of crime clusters in the City of Chicago with 2010-2020 crime trends.
Chicago Crime Clustering and Trends 
Overview
In this project, we have developed an interactive presentation of crime data in the city of Chicago for the period of 2010 to 2020. This project was designed to showcase two novel techniques:
1. Use of VRAM (GPU memory) and an embarrassingly parallel approach for processing big data using RAPIDS.
2. GMM clustering accelerated by a K-Means hybrid step.
Demo
Basic Functionality
Prerequisites
Data
Please ensure the data prerequisites are met by following the instructions presented in the README.md file here.
Tokens
Please obtain Google API and Mapbox API keys and place them in their appropriate files *-token.txt within the tokens folder.
Environment
Run the following code in the terminal and use the chicago kernel to run the main.ipynb notebook:
conda create -n chicago -c conda-forge python==3.8 -y;
conda activate chicago;
pip install -r requirements.txt
As long as you have the nb_conda_kernels package installed in your Jupyter environment, you should be able to access the chicago kernel inside Jupyter. We recommend using JupyterLab instead of Jupyter Notebook.