Working on the https://respicast.ecdc.europa.eu/ platform we noticed a feature of the validations package that modelers don't find desirable.
When checking that the values for quantiles are decreasing, the validations give the following error if the quantiles are not ordered in the input file:
! 2024-01-10-ISI-FluABCaster.csv: Values in 'value' column are not non-decreasing as output_type_ids increase for all unique task ID value/output type combinations of quantile or cdf output types. Seeerror_tblattribute for details.
Once the quantiles are ordered the validation step is passed without errors:
✔ 2024-01-10-ISI-FluABCaster.csv: Values in 'value' column are non-decreasing as output_type_ids increase for all unique task ID value/output type combinations of quantile or cdf output types.
Is there a rationale for this? Would it be possible to perform the check without requiring that the records are written in the file ordered by quantile?
Attached you can find the ordered and unordered csv files used to reproduce this issue.
Working on the https://respicast.ecdc.europa.eu/ platform we noticed a feature of the validations package that modelers don't find desirable. When checking that the values for quantiles are decreasing, the validations give the following error if the quantiles are not ordered in the input file:
! 2024-01-10-ISI-FluABCaster.csv: Values in 'value' column are not non-decreasing as output_type_ids increase for all unique task ID value/output type combinations of quantile or cdf output types. See
error_tblattribute for details.
Once the quantiles are ordered the validation step is passed without errors:
✔ 2024-01-10-ISI-FluABCaster.csv: Values in 'value' column are non-decreasing as output_type_ids increase for all unique task ID value/output type combinations of quantile or cdf output types.
Is there a rationale for this? Would it be possible to perform the check without requiring that the records are written in the file ordered by quantile?
Attached you can find the ordered and unordered csv files used to reproduce this issue.
Many thanks for your kind assistance Paolo
2024-01-10-ISI-FluABCaster_not_ordered.csv 2024-01-10-ISI-FluABCaster_ordered.csv