Merge pull request #21 from takuti/mf-alias

Make `MF` alias of `MatrixFactorization` fo readability

src/Recommendation.jl

@@ -22,7 +22,7 @@ include("model/tf_idf.jl")
 include("model/user_knn.jl")
 include("model/item_knn.jl")
 include("model/svd.jl")
-include("model/mf.jl")
+include("model/matrix_factorization.jl")
 
 include("metric/base.jl")
 include("metric/accuracy.jl")

src/model/mf.jl → src/model/matrix_factorization.jl

@@ -1,7 +1,7 @@
-export MF
+export MatrixFactorization, MF
 
 """
-    MF(
+    MatrixFactorization(
         data::DataAccessor,
         k::Int
     )
@@ -16,13 +16,13 @@ MF solves the following minimization problem for a set of observed user-item int
 
 where ``\\mathbf{p}_u, \\mathbf{q}_i \\in \\mathbb{R}^k`` are respectively a factorized user and item vector, and ``\\lambda`` is a regularization parameter to avoid overfitting. An optimal solution will be found by stochastic gradient descent (SGD). Ultimately, we can predict missing values in ``R`` by just computing ``PQ^{\\mathrm{T}}``, and the prediction directly leads recommendation.
 """
-struct MF <: Recommender
+struct MatrixFactorization <: Recommender
     data::DataAccessor
     k::Int
     P::AbstractMatrix
     Q::AbstractMatrix
 
-    function MF(data::DataAccessor, k::Int)
+    function MatrixFactorization(data::DataAccessor, k::Int)
         n_user, n_item = size(data.R)
         P = matrix(n_user, k)
         Q = matrix(n_item, k)
@@ -31,6 +31,8 @@ struct MF <: Recommender
     end
 end
 
+const MF = MatrixFactorization
+
 MF(data::DataAccessor) = MF(data, 20)
 
 isbuilt(recommender::MF) = isfilled(recommender.P)

test/model/test_mf.jl → test/model/test_matrix_factorization.jl

@@ -1,17 +1,26 @@
-function test_mf()
-    println("-- Testing MF-based recommender")
-
+function run(recommender::Type{T}) where {T<:Recommender}
     m = [NaN 3 NaN 1 2 1 NaN 4
          1 2 NaN NaN 3 2 NaN 3
          NaN 2 3 3 NaN 5 NaN 1]
     data = DataAccessor(m)
 
-    recommender = MF(data, 2)
+    recommender = recommender(data, 2)
     build!(recommender, learning_rate=15e-4, max_iter=100)
 
     # top-4 recommantion list should be same as CF/SVD-based recommender
     rec = recommend(recommender, 1, 4, [i for i in 1:8])
     @test Set([first(r) for r in rec]) == Set([2, 5, 6, 8])
 end
 
+function test_mf()
+    println("-- Testing MF-based (aliased) recommender")
+    run(MF)
+end
+
+function test_matrix_factorization()
+    println("-- Testing Matrix Factorization-based recommender")
+    run(MatrixFactorization)
+end
+
 test_mf()
+test_matrix_factorization()

test/runtests.jl

@@ -15,7 +15,7 @@ include("model/test_tf_idf.jl")
 include("model/test_user_knn.jl")
 include("model/test_item_knn.jl")
 include("model/test_svd.jl")
-include("model/test_mf.jl")
+include("model/test_matrix_factorization.jl")
 
 include("metric/test_accuracy.jl")
 include("metric/test_ranking.jl")