cejkaluk / covid-19-monte-carlo-simulation Goto Github PK

• Implement humans getting some incubation period when they get sick.

Simulation object should encapsulate the simulation and keep track of any data that is needed.

• DNC (Daily New Cases)

• Summarize the theory from the referenced paper into a theory chapter

Currently, the axis tick label font differs from the font of other text fields in the figure. The other text fields all use the latex interpreter.

• Make axis tick label font use the latex interpreter

According to the paper, it seems that humans move 1/4*7.9m as mean distance and 1/12*7.9m as std_var.
But, the city in the model is 1000m*1000m, so the humans in that simulation would barely move anywhere -> barely any spread.

Maybe do they mean 1/4 distance of the entire city? i,.e. mean movement to be 250m?
Then humans in the simulation move much more, maybe too much.

• Ask Gaspar on teams.

• Present results from simulation.
• Discuss accuracy of results when compared to the referenced paper.
• For resulting figures, include #18.

• Find formula to calculate it.
• Calculate it from data
• Create plot that is similar to the one in the paper.

• Print out current configuration before run of simulation.
• Make code look presentable.
• Document code.

• Move infect_susceptible_function to human class
• Rename infected to infectious
• Move config and any functions related to it to a separate class
• Move data and any function related to it to a separate class

Currently, humans have a status parameter, but no setting.

• Implement humans getting sick from other humans in their vicinity given a distance (random variable - see paper).

Currently, x and y coordinates are both generated from normal distribution with some mean and variance.

• Implement Cylindrical coordinates, where the direction will be from a uniform distribution and the step size from a normal distribution with specific length (see paper).

• Implement initialization of city with its humans equally space apart.

• MATLAB, the code, division into classes, capturing results, resulting figures.
• Mention differences between implementation and referenced paper.
• For differences include #7.

• Results: Not sure where the authors got their R_e. At first they have R_e = 3 => every infected person spread infected 3 other susceptible people. This would mean that from 10 cases -> 30 in one day (their DNC shows from 10 to 30 in roughly 10 days). I was able to achieve same R_e only when I used 3*8m as mean infection distance.
• The stuff about R_e filtering - authors used quite a lot of it and they didn't specify how they calculated the error bars.

• Make humans unable to leave the city bounds - if they try, have them bounce off or change direction.

• Create a header
• Release version 1.0