You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When calculating vincent ensembles, if the projection ever reports the same value for all quantiles in a specific grouping, the vincent function throws the error "'quant_vals' not found."
In the files attached, there is one subset that triggers this error in vincent (several of the groupings report 1 unique value for all quantiles) and another subset that does not trigger the error (all groups report more than 1 unique value for all quantiles).
And using the subsequent code:
`data <- read_excel("Vincente_issue_example.xlsx")
####version of quant_old that does not work in Vincente function
quant_old_bad<-quant_old_example %>%filter(model=="A" | model=="B")
###in quant_old_bad, there are instances where all the quantiles for one location, model, age_group report the same value. In quant_old_bad_unique this is represented when the last column has a value=1
quant_old_bad_unique<-quant_old_bad%>%group_by(model, location_name.x,age_group, population.y, scenario_id, value_r)%>%summarise(unique.value=length(unique(quantile.value)))
vincent function not working. throwing error:"quant_vals not found"
####Following subset does work in vincente function:
####version of quant_old that does not work:
quant_old_good<-quant_old_example%>%filter(age_group=="0-130" &(model=="C" | model=="B"))
###in quant_old_good, there is high variability in quantiles
quant_old_good_unique<-quant_old_good%>%group_by(model, location_name.x,age_group, population.y, scenario_id, value_r)%>%summarise(unique.value=length(unique(quantile.value)))
When calculating vincent ensembles, if the projection ever reports the same value for all quantiles in a specific grouping, the vincent function throws the error "'quant_vals' not found."
In the files attached, there is one subset that triggers this error in vincent (several of the groupings report 1 unique value for all quantiles) and another subset that does not trigger the error (all groups report more than 1 unique value for all quantiles).
Using this file to illustrate the issue:
ExcelVincente_issue_example.xlsx
And using the subsequent code:
`data <- read_excel("Vincente_issue_example.xlsx")
####version of quant_old that does not work in Vincente function
quant_old_bad<-quant_old_example %>%filter(model=="A" | model=="B")
###in quant_old_bad, there are instances where all the quantiles for one location, model, age_group report the same value. In quant_old_bad_unique this is represented when the last column has a value=1
quant_old_bad_unique<-quant_old_bad%>%group_by(model, location_name.x,age_group, population.y, scenario_id, value_r)%>%summarise(unique.value=length(unique(quantile.value)))
vincent function not working. throwing error:"quant_vals not found"
vincent_not_working=aggregate_cdfs(data=quant_old_bad %>% dplyr::rename(value=quantile.value),
id_var = "model",
group_by = c("location_name.x", "age_group", "population.y",
"scenario_id","value_r"),
method = "vincent",
ret_quantiles = quantile.vec)
####Following subset does work in vincente function:
####version of quant_old that does not work:
quant_old_good<-quant_old_example%>%filter(age_group=="0-130" &(model=="C" | model=="B"))
###in quant_old_good, there is high variability in quantiles
quant_old_good_unique<-quant_old_good%>%group_by(model, location_name.x,age_group, population.y, scenario_id, value_r)%>%summarise(unique.value=length(unique(quantile.value)))
vincent function works here
vincent_working=aggregate_cdfs(data=quant_old_good %>% dplyr::rename(value=quantile.value),
id_var = "model",
group_by = c("location_name.x", "age_group", "population.y",
"scenario_id","value_r"),
method = "vincent",
ret_quantiles = quantile.vec)`
The text was updated successfully, but these errors were encountered: