Repository navigation
Expand file tree
/
Copy pathMain.scala
More file actions
103 lines (81 loc) · 3.03 KB
/
Copy pathMain.scala
File metadata and controls
103 lines (81 loc) · 3.03 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
import function._
import scala.util.Random
import scala.math
object Main {
def main(args: Array[String]){
//挑出每一個user為候選
//val userCandidate = List.range(1, numUsers+1)
//挑出所有有兩份評分的user為候選 (取一份評分當test後，至少還留下一份做train)
//val userCandidate = ratingFile.map{ _.userID }.toSet.toList
val userCandidate = ratingFile.map{ _.userID }
.groupBy(x => x)
.filter{case(x,y) => y.size > 2}
.map{case(x,y) => x}
.toList
//隨機從候選users中挑出testSize個test users
val testSize = (0.3 * userCandidate.size).toInt
//從有n個候選user的candidate中隨機選出count個test users
def generateTestUsers(candidate: List[Int], n: Int, count: Int): List[Int] = {
if(count == 0)
Nil
else{
val i = Random.nextInt(n)
candidate(i) :: generateTestUsers((candidate.take(i) ::: candidate.drop(i+1)), n-1, count-1)
}
}
//儲存 test users 的資料，並將預計測試的電影在matrix重設為0
val testData = generateTestUsers(userCandidate, userCandidate.size, testSize).map{ id =>
val recent = ratingFile
.filter( _.userID == id )
.reduceLeft( (a,b) => if (a.timestamp > b.timestamp) a else b)
val actualRating = ratings(id - 1)(recent.movieID - 1)
ratings(id - 1)(recent.movieID - 1) = 0.0
RatingDataStructure(id, recent.movieID, actualRating, recent.timestamp )
}
//Training
val matrix = selectAlgorithm match {
case 1 =>
println("Running matrix fatorization algorithm")
println
new MatrixFacotrization
case 2 =>
println("Running SVD algorithm")
println
new SVD
case 3 =>
println("Running SVD++ algorithm")
println
new SVDplus
case 4 =>
println("Running timeSVD algorithm")
println
new TimeSVD
case 5 =>
println("Running timeSVD++ algorithm")
println
new TimeSVDplus
}
var mae: Double = 0.0
var rmse: Double = 0.0
var evaluateCount: Int = 0
//Test
for(test <- testData){
val actualRating = test.rating
//val predictRating = matrix.predict(test.userID-1, test.movieID - 1)
val predictRating = matrix.scalePredict(test.userID-1, test.movieID - 1)
println("User " + test.userID + " with movie " + test.movieID + " : ")
println(" Predic rating " + "%.3f".format(predictRating) )
println(" Actual rating " + test.rating)
println
mae += math.abs(actualRating - predictRating)
rmse += (actualRating - predictRating) * (actualRating - predictRating)
evaluateCount += 1
}
println("File name : " + filePath)
println("There are " + ratingFile.size + " ratings from " + numUsers + " users on " + numMovies + " movies")
println("Number of test cases : " + testData.size)
println("Number of factors : " + numFactors)
println("MAE = " + "%.3f".format(mae / evaluateCount) )
println("RMSE = " + "%.3f".format(math.sqrt(rmse / evaluateCount)) )
} //end of def main()
} // end of object SVD