Remove Missing Data

rm_missing(input: any[] | Mat | Tensor, dim: Number) : any[] | Mat | Tensor

param input - the input to remove the missing values of

param dim '0' or '1' : 0 means delete the whole column, '1' means delete the whole row. Default is '0'.

returns: any[] | Mat | Tensor - input without the missing values.

This function simply takes the input, and deletes the columns (default) or rows (if you set the parameter dim=1) that have missing values. This evidently deletes non-missing values in most cases.

Note that null, NaN, '', undefined are defined to be missing values. [] and {} are not.

This is useful when you simply want to remove missing data values and not handle it. This can perhaps be appropriate for very large arrays where the missing values are sparse and you dont want to fill it with false values.