DarkGauss/LSTMforCovidData

Uses a lstm to predict covid results on a merged dataset

LSTMforCovidData

Uses a lstm to predict covid results on a merged dataset

Note: In order to run the data processing scripts, the user must acquire the files COVID-19_Cases.csv, available from Johns Hopkins (link below) as it is too large for the repository.

• Google COVID 19 mobility reports:
  • https://www.google.com/covid19/mobility/
• Global Covid-19 Data (Johns Hopkins - cases):
  • https://data.world/covid-19-data-resource-hub/covid-19-case-counts
• The COVID Tracking Project (testing data by state):
  • https://github.com/COVID19Tracking/covid-tracking-data/blob/master/data/states_daily_4pm_et.csv
• United States by Density 2020
  • https://worldpopulationreview.com/states/state-densities/

In order to generate the data used in the project, the dataChanger.py script should be run first on the files it specifies. Once this is done, run dataJoin.py with an empty 'output' directory at its same directory level, along with the four files generated by dataChanger.py.

Requirements

• Python 3.5.x
• Jupyter Notebook 6.0.x
• TensorFlow 2.x.x
• seaborn 0.10.x
• pydot 1.4.x
• pandas 1.0.3
• numpy 1.18.x
• graphviz 0.14
• scikit-learn 0.22.x
• matplotlib 3.2.x