/// <summary>
/// Randomly splits this DataFrame with the provided weights.
/// Reference to https://github.com/apache/spark/blob/branch-1.4/python/pyspark/sql/dataframe.py, randomSplit(self, weights, seed=None)
/// </summary>
/// <param name="weights">list of weights with which to split the DataFrame. Weights will be normalized if they don't sum up to 1.0</param>
/// <param name="seed">The seed for sampling</param>
/// <returns></returns>
public IEnumerable <DataFrame> RandomSplit(IEnumerable <double> weights, int?seed = null)
{
foreach (var weight in weights)
{
if (weight < 0.0)
{
throw new ArgumentException(string.Format("Weights must be positive. Found weight value: {0}", weight));
}
}
if (seed == null)
{
seed = new Random().Next();
}
return(dataFrameProxy.RandomSplit(weights, seed.Value).Select(dfProxy => new DataFrame(dfProxy, sparkContext)));
}
