Orkin Pest Control recently named Washington DC the third "rattiest" city in America, and long-time DC residents know that rodent populations have been on the rise. Mayor Bowser announced a "Rat Riddance" intitiative in 2016, which aims to reduce rodent populations through changes in commercial practice and community awareness campaigns. In support of these efforts, the DC Office of the Chief Technology Officer (OCTO) has been analyzing 311 service request data related to rodent abatement requests and developing models to predict upticks in rat-related complaints in space and time.
The goals for our project are 2-fold:
- Develop models that can predict long-term trends in rodent complaints. These models will be complementary to ongoing OCTO efforts.
- Build out features on a public-facing 311 data portal that allows users to examine trends in service complaints over time and in their neighborhoods. This data portal will encompass all nuisance-related service complains, not just those related to rodents.
- Check out the "rats" channel on the Code for DC slack!
- Access our prototype of the 311 data portal here
- Download the 311 data from Dropbox here
- View our open issues on Trello here
We have rolled some starting R code and 311 data into a virtual image that can be accessed using a software called Docker. This allows users to run and modify R code that explores the 2016 311 data without needing to install R and script-dependent R packages.
- Get started by installing the free Docker Community Edition for your operating system:
- download for Windows 10 Professional (additional installation instructions)
- download for Mac (additional installation instructions)
- follow instructions for installation in Ubuntu
- Next, run the Docker image for our project from Code for DC's Docker Hub. Open a Terminal or shell and use the following command (Note that you will need to enter your administrator/sudo password into the command line after executing the command):
$ sudo docker run -d -p 8787:8787 codefordc2/explore-311-data-in-r
If you were unable to pull the Docker image from Docker Hub, try downloading the Docker image for our project, called "RStudio Server Demo.zip" from this Dropbox folder. Then, build the Docker image with the following command:
$ sudo docker build -t codefordc2/explore-311-data-in-r .
Next, launch the server with the following command:
$ sudo docker run -d -p 8787:8787 codefordc2/explore-311-data-in-r
- Finally, navigate to localhost:8787 in your favorite browser by pasting "localhost:8787" into the URL bar. If prompted to log in to the RStudio instance, the username and password are both 'rstudio'. You should now see a running version of RStudio with scripts and data related to the 311 data. Now you can run the scripts, view figures, and edit the code in the browser as if you had RStudio running locally on your computer.
Start by forking the repository, and then cloning the forked version of the repository to your computer. We use a triangular workflow - you should push to your fork, but fetch/pull from the Code for DC repo. Setting this up is easy. Use these commands:
$ git clone <url-of-your-fork>
$ cd the-rat-hack
$ git remote add codefordc https://github.com/codefordc/the-rat-hack.git
$ git remote -v
#you should see this:
codefordc https://github.com/codefordc/the-rat-hack.git (fetch)
codefordc https://github.com/codefordc/the-rat-hack.git (push)
origin <your/forked/url> (push)
origin <your/forked/url> (fetch)
Now instead of plain git push and git pull, use these:
$ git push origin <branch-name> #pushes to your forked repo
$ git pull codefordc <branch-name> #fetches and merges from the codefordc repo
Here’s more information on setting up triangular workflows (scroll to “Improved support…”).
Never worked with a triangular workflow before? Ask a project lead for help.