I have a PySpark dataframe-
df = spark.createDataFrame([ ("u1", [\'u1_row1\', \'u1_row2\', \'u1_row3\']), ("u2", [\'u2_row1\', \'u