runDemo.sh

echo -e "MLlib Matrix Factorization Demo\n"
echo -e "Please rate the following 20 movies from 1 to 5. If you haven't heard of the movie, enter a 0\n"

echo -e "MOVIE                     RATING"
read -p "Toy Story:                  " a
read -p "Jumanji:                    " b
read -p "Heat:                       " c
read -p "GoldenEye:                  " d
read -p "The American President:     " e
read -p "Nixon:                      " f
read -p "Sense and Sensibility:      " g
read -p "Othello:                    " h
read -p "Babe:                       " i
read -p "Clueless:                   " j
read -p "Mortal Kombat:              " k
read -p "Pochahontas:                " l
read -p "The Usual Suspects:         " m
read -p "Les Miserables:             " n
read -p "Apollo 13:                  " o
read -p "Batman Forever:             " p
read -p "Die Hard: With a Vengeance: " q
read -p "Free Willy 2:               " r
read -p "Mallrats:                   " s
read -p "Pulp Fiction:               " t


echo -e "1,69879,"$a "\n2,69879,"$b "\n6,69879,"$c "\n10,69879,"$d "\n11,69879,"$e "\n14,69879,"$f "\n17,69879,"$g "\n26,69879,"$h "\n34,69879,"$i "\n39,69879,"$j "\n44,69879,"$k "\n48,69879,"$l "\n50,69879,"$m "\n73,69879,"$n "\n150,69879,"$o "\n153,69879,"$p "\n165,69879,"$q "\n169,69879,"$r "\n180,69879,"$s "\n296,69879,"$t >> data/movielens.data

echo "Thanks!"

eccho "Adding your data ..."

~/ephemeral-hdfs/bin/hadoop fs -put data /data

echo "Running matrix factorization"

../spark/run-example org.apache.spark.mllib.recommendation.ALS spark://ec2-54-237-80-146.compute-1.amazonaws.com:7077 hdfs://ec2-54-237-80-146.compute-1.amazonaws.com:9000/data/movielens.data 10 10 /results

echo "Finding recommended movies"

sbt/sbt "run-main moviedemo.MovieRec --master=spark://ec2-54-237-80-146.compute-1.amazonaws.com:7077 --moviefeats=hdfs://ec2-54-237-80-146.compute-1.amazonaws.com:9000/results/userFeatures --userfeats=hdfs://ec2-54-237-80-146.compute-1.amazonaws.com:9000/results/productFeatures --sparkhome=/root/spark --jar=/root/moviedemo/target/scala-2.9.3/moviedemo-assembly-1.0.jar --userid=69879 --titles=hdfs://ec2-54-237-80-146.compute-1.amazonaws.com:9000/data/movies.dat"

echo "Erasing old results.."

~/ephemeral-hdfs/bin/hadoop fs -rmr hdfs://ec2-54-237-80-146.compute-1.amazonaws.com:9000/data/
~/ephemeral-hdfs/bin/hadoop fs -rmr hdfs://ec2-54-237-80-146.compute-1.amazonaws.com:9000/results/
cp ../movielens.data.o data/movielens.data