Spearman correlation per groups in pyspark

Question

Is there any way to get a correlation per grouped data?

For example

Country  | rank1 | rank2
Us         3       4
Us         1       2
Us         4       1