GroupBy

GroupBy is a useful tool for data grouping and aggregation in DolphinDB pandas. With GroupBy, users can group data based on a single column or more, and apply a function (such as aggregation, transformation, or filtering) to each group independently.

Note:

  • All group-by columns must be of the same data type.

  • The GroupBy calculation results of all functions (other than cumcount) are sorted in a different order from the original input. The order within each group remains the same as the original, while that across groups is determined by the GroupBy column(s).

Function Application

Function

Compatibility Statement
apply

Computations / Descriptive Statistics

Function

Compatibility Statement
rolling Only parameters window and min_periods are supported.
cumprod No parameter is supported.
cumsum No parameter is supported.
diff No parameter is supported.
max Only the parameter numeric_only is supported.
min Only the parameter numeric_only is supported.
mean Only the parameter numeric_only is supported.
std Only the parameter numeric_only is supported.
sum The sum of literal data or data of POINT, UUID, IPADDR, INT128, or COMPLEX type returns NULL.
shift The parameter axis is not supported.
var Only the parameter numeric_only is supported.