This code is based on the example wordcount5 of the delivery package clojure-hadoop.

Assembly with leiningen:
    
    lein deps
    lein compile
    lein uberjar

Run in standalone mode:
    
    java -cp clojure_hadoop-1.0-SNAPSHOT-standalone.jar clojure_hadoop.job -job wordcount/job -input FILE -output out
    
FILE as an argument, replace the text file, and after the program out in the catalog file is created containing a list of N-Gram and their number in the text.

Run in Hadoop cluster:
    
    hadoop jar clojure_hadoop-1.0-SNAPSHOT-standalone.jar clojure_hadoop.job -job wordcount/job -input input -output output
    
The files should be placed on HDFS