'validate_data' checks the data for expected columns and data types, removes unneeded columns, and returns the resulting data. Missing or NA values will be removed with a warning. Valid data must include the following columns (and types):
patient_mmwr_year
(integer)patient_mmwr_week
(integer)patient_disease
(character)
Examples
df <- data.frame(
patient_mmwr_year = 2020L,
patient_mmwr_week = 1L,
patient_disease = "A"
)
validate_data(df)
#> patient_mmwr_year patient_mmwr_week patient_disease
#> 1 2020 1 A