The epiworldr from uofuepibio

epiworldr's Issues

Write a wrapper of the `models` namespace

The epiworld C++ library has many predefined models. These should be included directly via R wrappers. The list of models to add is:

This list will grow, eventually. I'll start with the first one.

SIGSEGV in ModelSEIR when removing virus

I am getting the following error when running playground/benchmark-seir.R using R -gdb:

Program received signal SIGSEGV, Segmentation fault.
0x00007ffff4466fcc in epiworld::default_rm_virus<int> (a=..., model=0x555558435ee0) at ../inst/include/epiworld/agent-actions-meat.hpp:114
114         model->get_db().today_virus[v->get_id()][p->state_prev]--;

This happens on the speedup PR.

Specifications of first infected in ModelDiffNet

I was making a model for opinion diffusion with ModelDiffNet(), where each node has the following attribute: a real number beta between 0 and 1, following a certain distribution. But, for the model, its important that the first infecterd meet the condition beta<0.2. This restriction of selection in early infected individuals cannot be accomplished with the current packege in R, epiworldR. Can you adapt the C++ code to allow this kind of escenarios?

Sunbelt issues

Set a unique id for each run in run_multiple.
OpenMP not working properly with a high number of threads.
Review the ModelDiffNet module; probabilities are not as expected.

Improve docs in plot functions

plot_reproductive_number(...) ellipsis is not clear.
Same happens to the plot_generation_time(...).
Both need documentation regarding what the (invisible) output is.

Reported by @AnibalLuciano.

plot: Draw a figure showing the counts (y-axis) throughout the dates (x-axis). It should be similar to what it is implemented right now here. In particular, we will need one method for each class like we do here:

#' @rdname ModelSEIR
#' @export
plot.epiworld_seir <- function(x, y = NULL, ...) {
  ...[do magic]...
}

plot_r: Plot the overall reproductive number (see here). This should have a new generic, since the function plot_r does not exist.

Higher contact rate error

result <- suppressWarnings(suppressMessages(invisible({
  
  library(epiworldR)
  
  library(data.table)
  library(EpiEstim)
  library(dplyr)
  library(ggplot2)
  library(tidyverse)
  library(EpiNow2)
})))


model_seircon=ModelSEIRCONN(name="covid",n=50000,prevalence = 0.001,contact_rate = 20,transmission_rate = 0.5,recovery_rate = 1/7,incubation_days = 3)





run(model_seircon,ndays=50,seed=1912)
#> _________________________________________________________________________
#> |Running the model...
#> |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| done.
#> | done.
summary(model_seircon)
#> ________________________________________________________________________________
#> ________________________________________________________________________________
#> SIMULATION STUDY
#> 
#> Name of the model   : Susceptible-Exposed-Infected-Removed (SEIR) (connected)
#> Population size     : 50000
#> Agents' data        : (none)
#> Number of entities  : 0
#> Days (duration)     : 50 (of 50)
#> Number of viruses   : 1
#> Last run elapsed t  : 473.00ms
#> Last run speed      : 5.28 million agents x day / second
#> Rewiring            : off
#> 
#> Global events:
#>  (none)
#> 
#> Virus(es):
#>  - covid (baseline prevalence: 0.10%)
#> 
#> Tool(s):
#>  (none)
#> 
#> Model parameters:
#>  - Avg. Incubation days : 3.0000
#>  - Contact rate         : 20.0000
#>  - Prob. Recovery       : 0.1429
#>  - Prob. Transmission   : 0.5000
#> 
#> Distribution of the population at time 50:
#>   - (0) Susceptible : 49950 -> 0
#>   - (1) Exposed     :    50 -> 0
#>   - (2) Infected    :     0 -> 140
#>   - (3) Recovered   :     0 -> 49860
#> 
#> Transition Probabilities:
#>  - Susceptible  0.60  0.40  0.00  0.00
#>  - Exposed      0.00  0.66  0.34  0.00
#>  - Infected     0.00  0.00  0.86  0.14
#>  - Recovered    0.00  0.00  0.00  1.00



plot(model_seircon)



saver <- make_saver(
  "total_hist",
  "transmission",
  "transition",
  "reproductive",
  "generation"
)
run_multiple(model_seircon,ndays=50,nsim=100,seed=1972,saver=saver)
#> Starting multiple runs (100) using 1 thread(s)
#> _________________________________________________________________________
#> _________________________________________________________________________
#> ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| done.
#>  done.


res1 <- run_multiple_get_results(model_seircon)
#> Warning in run_multiple_get_results(model_seircon): When retrieving the saved
#> results, for the case of transmission, there were no observations.
#> Error in attributes(.Data) <- c(attributes(.Data), attrib): all attributes must have names [4 does not]
res1$reproductive
#> Error in eval(expr, envir, enclos): object 'res1' not found

First steps

Create a skeleton package (empty) using either Rcpp or cpp11 (describe the reasoning behind your decision.)
Upload the package to GitHub.
Create a README.Rmd file for the package. Use the description of https://github.com/UofUEpi/epiworld.

uofuepibio / epiworldr Goto Github PK

epiworldr's People

Contributors

Stargazers

Watchers

Forkers

epiworldr's Issues

Write a wrapper of the `models` namespace

SIGSEGV in ModelSEIR when removing virus

Specifications of first infected in ModelDiffNet

Sunbelt issues

Improve docs in plot functions

epiworldR compiling

Documentation for generation time

Methods

Higher contact rate error

First steps

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent