how to set missing time series values to 0 in pyspark dataframe

Question

I have a dataset given below:

product_name    serial_number     date           sum
"A"             "12"              "2020-01-01"