Social Distancing

This repository contains an epidemic simulation outcome under two different strategies:

• Normal social interactions
• Social distancing

Setup

1. Create and activate a virtual environment:
   • Create: python -m virtualenv --python=python3.7 venv
   • Activate Linux/MacOS: source venv/bin/activate
   • Activate Windows: source venv/Scripts/activate
2. Install requirements:
   • pip install -r requirements.txt

The application will run according to the environment variables you define. If a particular config is not found on the environment, either the default.json values will be used or a hard-coded value on the settings.py module.

All environment variables use the SOCIAL_DISTANCING_ prefix since that's the general application name. Relevant environment variables:

• SOCIAL_DISTANCING_VAR_POPULATION
  • Description: Number of people to simulate.
  • Default: 1000
• SOCIAL_DISTANCING_VAR_STRATEGY
  • Description: Strategy to be used during simulation (determines default values).
  • Options: social-interacting, social-distancing
  • Default: Determined on default.json.
• SOCIAL_DISTANCING_VAR_PERCENT_STUDENTS
  • Description: Number of students per 100 people.
  • Default: Determined on default.json.
• SOCIAL_DISTANCING_VAR_PERCENT_WORKERS
  • Description: Number of workers per 100 people.
  • Default: Determined on default.json.
• SOCIAL_DISTANCING_VAR_PERCENT_WORKER_STUDENTS
  • Description: Number of worker-students per 100 people.
  • Default: Determined on default.json.
• SOCIAL_DISTANCING_VAR_PERCENT_STAY_HOME
  • Description: Number of stay-at-home people per 100 people.
  • Default: Determined on default.json.
• SOCIAL_DISTANCING_VAR_FACTORY_NUM_HOMES
  • Description: Number of homes per 100 people.
  • Default: 40
• SOCIAL_DISTANCING_VAR_FACTORY_NUM_WORKPLACE
  • Description: Number of workplaces per 100 people.
  • Default: 10
• SOCIAL_DISTANCING_VAR_FACTORY_NUM_UNIVERSITY
  • Description: Number of universities (schools) per 100 people.
  • Default: 1
• SOCIAL_DISTANCING_VAR_FACTORY_NUM_PUBLIC
  • Description: Number of public places per 100 people.
  • Default: 15

Use case

Open 2 terminals, activate the virtualenv and run the following commands (one per terminal):

$ SOCIAL_DISTANCING_VAR_STRATEGY="social-interacting" python main.py simulate-multiple --name out-1 --simulations 5 --days 300

$ SOCIAL_DISTANCING_VAR_STRATEGY="social-distancing" python main.py simulate-multiple --name out-1 --simulations 5 --days 300

The output of the social-interacting and social-distancing strategies will be combined on the same directory named out-1.

Now you can run the analyze command:

python main.py analyze --simulation-name out-1 --days 250 --show

Usage

Run the application as following:

$ python main.py [command] [options]

You can specify a particular environment variable on the context of the application as following:

$ SOCIAL_DISTANCING_VAR_STRATEGY="social-distancing" python main.py [command] [options]

The environment variable ALWAYS overwrites the default.json values.

Show strategies

Print the strategies on the commandline.

show-strategies

Example

$ python main.py show-strategies
* social-interacting
* social-distancing

Simulate

Simulate a single scenario:

simulate --days {days} --filename {filename} --show

    --days: integer representing the number of days to simulate.
    --filename: if present, saves the simulation results output on a csv-file.
    --show: if present, shows a plot of the confirmed cases over time.

Example

Run a single simulation of the pandemic outcome during 300 days using the social distancing strategy and show the results.

SOCIAL_DISTANCING_VAR_STRATEGY="social-distancing" python main.py simulate --days 300 --show

Simulate Multiple (Recommended)

Run multiple simulations on parallel.

simulate-multiple --name {name} --simulations {n} --days {days}

    --name: string representing the simulation name. A directory will be created containing the simulation(s) output.
    --simulations: integer representing the number of simulations to run. 
    --days: integer representing the number of days to simulate.
    

Example

Run 5 simulations of the pandemic output for 300 days using the social distancing strategy. Save the results on a directory called out-1.

SOCIAL_DISTANCING_VAR_STRATEGY="social-distancing" python main.py simulate-multiple --name out-1 --simulations 5 --days 300

Analyze (Recommended)

Analyze the aggregate results of the Simulate Multiple command.

analyze --simulation-name {name} --days {days} --avg-days {avg-days} --save {filename} --show 

    --simulation-name: string representing the path of the multiple-simualtion results.
    --days: number of max days to analyze. Default value is `None` meaning to not filter out days.
    --avg-days: Plot the average of n-days. Default: 10.
    --save: filename (png) to save the plot.
    --show: show plot. 

Example

Analyze the first 200 days of the out-1 simulation:

python main.py analyze --simulation-name out-1 --days 200 --show