Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

generate_initial_disease_states() errors when number_initial_E = 0, and can miscount when number_initial_S = 0

Aperta Adatta ai principianti
  • Qualcosa non va?
#161 0 commenti 0 reazioni 0 assegnatari Vedi su GitHub

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
2/5
Tempo stimato
1-3 ore
Idoneità per principianti
92/100
Tipo di issue
Bug
Chiarezza
Specificata chiaramente
Stato di attività
Attiva
Stack tecnologico
r
Ambito
backend

Direzione di ricerca

Inizia con generate_initial_disease_states() e riproduci entrambi gli errori usando gli esempi di parametri presenti nell'issue. Aggiungi casi di regressione a test-variables.R per zero persone esposte e zero persone suscettibili, poi esegui i test R pertinenti. Il lavoro è completo quando ogni combinazione valida dei conteggi iniziali produce i conteggi S, E, I e R richiesti, inclusi i pool contenenti una sola persona.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

bug

Summary

generate_initial_disease_states() breaks on two kinds of valid initial conditions:

  1. number_initial_E = 0 with any initial I or R people errors with invalid first argument. For example, you can't start an epidemic from infectious people alone.
  2. number_initial_S = 0 can silently give the wrong counts. When only one person is left for the last draw, that draw can pick someone already assigned to an earlier state.

Both bugs are on main. They have been there since 6b5daf8 (2024-07-23), which added explicit initial I and R counts.

Reproducer

p <- list(human_population = 100, number_initial_S = 95, number_initial_E = 0,
          number_initial_I = 5, number_initial_R = 0, seed = 1)
generate_initial_disease_states(p)
#> Error in sample.int(length(x), size, replace, prob) : invalid first argument

The same error comes through get_parameters() and create_variables(), so run_simulation() can't start either.

p <- list(human_population = 10, number_initial_S = 0, number_initial_E = 9,
          number_initial_I = 1, number_initial_R = 0, seed = 1)
table(generate_initial_disease_states(p))
#>      E I_mild      S
#>      8      1      1

Seeds 1, 4 and 5 give E = 8 and S = 1 instead of E = 9 and S = 0. Seeds 2 and 3 happen to be correct.

Which inputs are affected

Tested with human_population = 100, with each count set to 0 or 5:

E I R Result
0 0 0 ok
0 > 0 any error
0 0 > 0 error
> 0 any any ok, unless S = 0 (below)

With S = 0, the counts are wrong for some seeds whenever exactly one person is left for the I draw or the R draw. For example, with N = 10: E = 9, I = 1, R = 0, or E = 5, I = 4, R = 1.

Cause

Two R gotchas in how the remaining pool is built and sampled:

  1. x[-integer(0)] returns an empty vector, not x. When E = 0, exposed_index is integer(0). So population_index[-exposed_index] is empty, and the I and R draws sample from nobody.
  2. sample(x, size) treats a single number x as 1:x. When one person is left in the pool, sample() draws from everyone up to that person's index, which can include people already made E or I. The draw overwrites them, and the person who should have been drawn stays S.

Proposed fix

Build each remaining pool with setdiff(), and draw with sample.int() so a pool of one person is handled correctly:

draw <- function(pool, size) pool[sample.int(length(pool), size)]

pool <- seq_along(initial_disease_states)
exposed_index <- draw(pool, parameters_list$number_initial_E)
initial_disease_states[exposed_index] <- "E"
pool <- setdiff(pool, exposed_index)

infectious_index <- draw(pool, parameters_list$number_initial_I)
initial_disease_states[infectious_index] <- "I_mild"
pool <- setdiff(pool, infectious_index)

recovered_index <- draw(pool, parameters_list$number_initial_R)
initial_disease_states[recovered_index] <- "R"

I checked this against the current function for every case above, with seeds 1 to 20:

  • It gives the right counts in every case.
  • It gives identical output in every case where the current function is already correct, so existing seeded runs don't change.

A single shuffle, sample(rep(c("S", "E", "I_mild", "R"), counts)), would be simpler. But it changes the initial state of every seeded run.

test-variables.R should get cases for E = 0 with I > 0, and for S = 0.

Lingua principale
R
Stelle
8
Fork
4
Nessuna PR unita negli ultimi 30g

Preparare l'ambiente

Questo progetto non fornisce container di sviluppo, Dockerfile né guida per i contributori, quindi l'ambiente è a tuo carico: parti dal suo README e consulta la nostra guida al primo contributo per i passaggi generali.

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di petal-code/helios

Tutte le issue di petal-code/helios

Issue simili

Altre issue su R

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.