/// <summary> /// Returns a new DataFrame by sampling a fraction of rows. /// </summary> /// <param name="withReplacement"> Sample with replacement or not. </param> /// <param name="fraction"> Fraction of rows to generate. </param> /// <param name="seed"> Seed for sampling. If it is not present, a randome long value will be assigned. </param> // Python API: https://github.com/apache/spark/blob/branch-1.4/python/pyspark/sql/dataframe.py // sample(self, withReplacement, fraction, seed=None) public DataFrame Sample(bool withReplacement, double fraction, long?seed) { long v; if (seed.HasValue) { v = seed.Value; } else { v = ((long)random.Next()) << 32 + random.Next(); } return(new DataFrame(dataFrameProxy.Sample(withReplacement, fraction, v), sparkContext)); }