U-SQL + Python基本问题

Question

我刚刚开始学习如何将Python与U-SQL集成。我正在做这个例子：

REFERENCE ASSEMBLY [ExtPython];

DECLARE @myScript = @"
def get_mentions(tweet):    
      return ';'.join( ( w[1:] for w in tweet.split() if w[0]=='@' ) )

def usqlml_main(df):    
      del df['time']    
      del df['author']    
      df['mentions'] = df.tweet.apply(get_mentions)    
      del df['tweet']    
      return df
";

@t  =     
SELECT * FROM        
(VALUES
           ("D1","T1","A1","@foo Hello World @bar"),
           ("D2","T2","A2","@baz Hello World @beer")
       ) AS
            D( date, time, author, tweet );
@m  =
    REDUCE @t ON date
    PRODUCE date string, mentions string
    USING new Extension.Python.Reducer(pyScript:@myScript);

OUTPUT @m
    TO "/tweetmentions.csv"
    USING Outputters.Csv();

在usqlml_main内部，df.tweet.apply(get_mentions)中的'apply‘函数是什么？

另外，REDUCE在这种情况下会做什么？在将U-SQL与Python集成时，总是需要这样做吗？