This project utilizes data from PetFinder.com, that was downloaded from kaggle. The goal is to predict the time it takes for dogs to get adopted based on the information pulled from PetFinder.com which contains characteristics like age, breed, color, vaccination status etc. This project makes use of R for data cleaning and visualization (dplyr, ggplot) and Python for building and testing machine learning models (PyCaret, Pandas).
To get started, clone this repository using the following command line argument in a bash shell
git clone https://github.com/skpeterson/Predicting_Pet_Adoptability.git
To build the docker container, run the following in the directory you cloned the repo
bash build_docker.sh
To run the Docker container, please run
bash start_docker.sh
The docker container runs both Rstudio and Python/Jupyter notebook.
- To access Rstudio, navigate to localhost:8787 in your browser. username: rstudio, password: benson
- To access Jupyter notebook go to http://<host_machine_ip>:8888/tree?token= . Where host machine IP is the IP address of the machine running the Docker container, and the token can be found in the container log in the bash shell you started the container in.There will be a link you can click to in the log that will take you to the Jupyter Notebook.
Great!! You're in!!
You're welcome to go exploring around, run a few scripts manually, generate some visualizations, or if you would just like to generate the final report, go to the terminal in the interactive Rstudio server session you joined and clean the directories so we can ensure we are generating new figures
make clean
initialize the directories we will need to store results
make init
and finally, build the report!
make /home/rstudio/work/results/report/Summary_Report.html