Skip to content

Commit a09dfb8

Browse files
filipebraidafilipebraida
committed
Formatting code
1 parent b6f16ea commit a09dfb8

File tree

3 files changed

+90
-90
lines changed

3 files changed

+90
-90
lines changed

‎src/movielens.jl

+15-15
Original file line numberDiff line numberDiff line change
@@ -1,15 +1,15 @@
11
const defdir = joinpath(dirname(@__FILE__), "..", "datasets")
22

33
function getmovielensdata(dir)
4-
mkpath(dir)
5-
path = download("http://files.grouplens.org/datasets/movielens/ml-100k.zip")
6-
run(unpack_cmd(path,dir,".zip", ""))
4+
mkpath(dir)
5+
path = download("http://files.grouplens.org/datasets/movielens/ml-100k.zip")
6+
run(unpack_cmd(path, dir, ".zip", ""))
77
end
88

99
function getmovielensdata1m(dir)
10-
mkpath(dir)
11-
path = download("http://files.grouplens.org/datasets/movielens/ml-1m.zip")
12-
run(unpack_cmd(path,dir,".zip", ""))
10+
mkpath(dir)
11+
path = download("http://files.grouplens.org/datasets/movielens/ml-1m.zip")
12+
run(unpack_cmd(path, dir, ".zip", ""))
1313
end
1414

1515
"""
@@ -18,15 +18,15 @@ end
1818
Return MovieLens 100k dataset.
1919
"""
2020
function MovieLens()::Persa.Dataset
21-
filename = "$(defdir)/ml-100k/u.data"
21+
filename = "$(defdir)/ml-100k/u.data"
2222

23-
isfile(filename) || getmovielensdata(defdir)
23+
isfile(filename) || getmovielensdata(defdir)
2424

25-
file = DataFrame(CSV.read(filename, delim = '\t',
25+
file = DataFrame(CSV.read(filename, delim = '\t',
2626
header = [:user, :item, :rating, :timestamp],
2727
types = [Int, Int, Int, Int]))
2828

29-
return Persa.Dataset(file)
29+
return Persa.Dataset(file)
3030
end
3131

3232
"""
@@ -45,11 +45,11 @@ function MovieLens1M()::Persa.Dataset
4545

4646
df = DataFrame()
4747

48-
df[!, :user] = convert(Array{Int}, file[!, :user])
49-
df[!, :item] = convert(Array{Int}, file[!, :item])
50-
df[!, :item] = labelencode(labelmap(df[!, :item]), df[!, :item])
51-
df[!, :rating] = convert(Array{Int}, file[!, :rating])
52-
df[!, :timestamp] = convert(Array{Int}, file[!, :timestamp])
48+
df[!, :user] = convert(Array{Int}, file[!, :user])
49+
df[!, :item] = convert(Array{Int}, file[!, :item])
50+
df[!, :item] = labelencode(labelmap(df[!, :item]), df[!, :item])
51+
df[!, :rating] = convert(Array{Int}, file[!, :rating])
52+
df[!, :timestamp] = convert(Array{Int}, file[!, :timestamp])
5353

5454
return Persa.Dataset(df)
5555
end

‎src/others.jl

+52-52
Original file line numberDiff line numberDiff line change
@@ -1,94 +1,94 @@
11
function Netflix()::Persa.CFDataset
2-
filename = "$(defdir)/netflix/netflix.csv"
3-
file = readtable(filename, separator = ',', header = false)
2+
filename = "$(defdir)/netflix/netflix.csv"
3+
file = readtable(filename, separator = ',', header = false)
44

55

6-
df = DataFrame()
6+
df = DataFrame()
77

8-
df[:user] = labelencode(labelmap(file[:,1]), file[:,1])
9-
df[:item] = file[:,2]
10-
df[:rating] = file[:,3]
8+
df[:user] = labelencode(labelmap(file[:,1]), file[:,1])
9+
df[:item] = file[:,2]
10+
df[:rating] = file[:,3]
1111

12-
return Persa.Dataset(df)
12+
return Persa.Dataset(df)
1313
end
1414

1515
function MovieTweeting()::Persa.TimeCFDataset
16-
filename = "$(defdir)/Movie-Tweeting-200k/ratings.dat"
17-
file = readtable(filename, separator = ':', header = false)
16+
filename = "$(defdir)/Movie-Tweeting-200k/ratings.dat"
17+
file = readtable(filename, separator = ':', header = false)
1818

19-
df = DataFrame()
19+
df = DataFrame()
2020

21-
df[:user] = file[:,1]
22-
df[:item] = labelencode(labelmap(file[:,2]), file[:,2])
23-
df[:rating] = file[:,3]
24-
df[:timestamp] = file[:,4]
21+
df[:user] = file[:,1]
22+
df[:item] = labelencode(labelmap(file[:,2]), file[:,2])
23+
df[:rating] = file[:,3]
24+
df[:timestamp] = file[:,4]
2525

26-
return Persa.Dataset(df)
26+
return Persa.Dataset(df)
2727
end
2828

2929
function MovieTweeting10k()::Persa.TimeCFDataset
30-
filename = "$(defdir)/mt-snapshot-10k/ratings.dat"
31-
file = readtable(filename, separator = ':', header = false)
30+
filename = "$(defdir)/mt-snapshot-10k/ratings.dat"
31+
file = readtable(filename, separator = ':', header = false)
3232

33-
df = DataFrame()
33+
df = DataFrame()
3434

35-
df[:user] = file[:,1]
36-
df[:item] = labelencode(labelmap(file[:,2]), file[:,2])
37-
df[:rating] = file[:,3]
38-
df[:timestamp] = file[:,4]
35+
df[:user] = file[:,1]
36+
df[:item] = labelencode(labelmap(file[:,2]), file[:,2])
37+
df[:rating] = file[:,3]
38+
df[:timestamp] = file[:,4]
3939

40-
return Persa.Dataset(df)
40+
return Persa.Dataset(df)
4141
end
4242

4343
function CiaoDVD()::Persa.TimeCFDataset
44-
filename = "$(defdir)/CiaoDVD/movie-ratings.txt"
45-
file = readtable(filename, separator = ',', header = false)
44+
filename = "$(defdir)/CiaoDVD/movie-ratings.txt"
45+
file = readtable(filename, separator = ',', header = false)
4646

47-
df = DataFrame()
47+
df = DataFrame()
4848

49-
df[:user] = file[:,1]
50-
df[:item] = file[:,2]
51-
df[:rating] = file[:,5]
52-
df[:timestamp] = convert(Array{Int}, Dates.datetime2unix.(Dates.DateTime(file[:,6])))
49+
df[:user] = file[:,1]
50+
df[:item] = file[:,2]
51+
df[:rating] = file[:,5]
52+
df[:timestamp] = convert(Array{Int}, Dates.datetime2unix.(Dates.DateTime(file[:,6])))
5353

54-
return Persa.Dataset(df)
54+
return Persa.Dataset(df)
5555
end
5656

5757
function FilmTrust()::Persa.CFDataset
58-
filename = "$(defdir)/FilmTrust/ratings.txt"
59-
file = readtable(filename, separator = ' ', header = false)
58+
filename = "$(defdir)/FilmTrust/ratings.txt"
59+
file = readtable(filename, separator = ' ', header = false)
6060

61-
df = DataFrame()
61+
df = DataFrame()
6262

63-
df[:user] = file[:,1]
64-
df[:item] = file[:,2]
65-
df[:rating] = file[:,3]
63+
df[:user] = file[:,1]
64+
df[:item] = file[:,2]
65+
df[:rating] = file[:,3]
6666

67-
return Persa.Dataset(df)
67+
return Persa.Dataset(df)
6868
end
6969

7070
function YahooMusic()::Persa.CFDataset
71-
filename = "$(defdir)/yahoo-music-r3/ymusic-r3-dummy-time.dat"
72-
file = readtable(filename, separator = ' ', header = false)
71+
filename = "$(defdir)/yahoo-music-r3/ymusic-r3-dummy-time.dat"
72+
file = readtable(filename, separator = ' ', header = false)
7373

74-
df = DataFrame()
74+
df = DataFrame()
7575

76-
df[:user] = file[:,1]
77-
df[:item] = file[:,2]
78-
df[:rating] = file[:,3]
76+
df[:user] = file[:,1]
77+
df[:item] = file[:,2]
78+
df[:rating] = file[:,3]
7979

80-
return Persa.Dataset(df)
80+
return Persa.Dataset(df)
8181
end
8282

8383
function LastFM()::Persa.CFDataset
84-
filename = "$(defdir)/lastfm/last_fm.dat"
85-
file = readtable(filename, separator = ',', header = false)
84+
filename = "$(defdir)/lastfm/last_fm.dat"
85+
file = readtable(filename, separator = ',', header = false)
8686

87-
df = DataFrame()
87+
df = DataFrame()
8888

89-
df[:user] = labelencode(labelmap(file[:,1]), file[:,1])
90-
df[:item] = labelencode(labelmap(file[:,2]), file[:,2])
91-
df[:rating] = file[:,3]
89+
df[:user] = labelencode(labelmap(file[:,1]), file[:,1])
90+
df[:item] = labelencode(labelmap(file[:,2]), file[:,2])
91+
df[:rating] = file[:,3]
9292

93-
return Persa.Dataset(df)
93+
return Persa.Dataset(df)
9494
end

‎test/analysis.jl

+23-23
Original file line numberDiff line numberDiff line change
@@ -52,39 +52,39 @@ factor = convert(Int, round(length(hist_user) / 200))
5252

5353
open("t.txt", "w") do f
5454
write(f, "n\tcount\n")
55-
for i=1:length(hist_user)
55+
for i = 1:length(hist_user)
5656
if i % factor == 1
5757
write(f, "$i\t$(hist_user[i])\n")
5858
end
5959
end
60-
end
60+
end
6161

6262
###
6363
factor = convert(Int, round(length(hist_item) / 200))
6464

6565
open("t.txt", "w") do f
6666
write(f, "n\tcount\n")
67-
for i=1:length(hist_item)
67+
for i = 1:length(hist_item)
6868
if i % factor == 1
6969
write(f, "$i\t$(hist_item[i])\n")
7070
end
7171
end
72-
end
72+
end
7373

7474
#####
7575
ds = datasets[3]()
7676
sort!(ds.file, cols = :timestamp)
7777
ds.file[:timestamp]
7878

79-
moments = Dict{Int, Int}()
79+
moments = Dict{Int,Int}()
8080

8181
times = unique(ds.file[:timestamp])
8282

83-
for i=1:length(times)
83+
for i = 1:length(times)
8484
moments[times[i]] = 0
8585
end
8686

87-
for i=1:length(ds.file[:timestamp])
87+
for i = 1:length(ds.file[:timestamp])
8888
moments[ds.file[:timestamp][i]] += 1
8989
end
9090

@@ -99,12 +99,12 @@ factor = convert(Int, round(size(moments2_sorted)[1] / 200))
9999

100100
open("t.txt", "w") do f
101101
write(f, "n\tcount\n")
102-
for i=1:size(moments2_sorted)[1]
102+
for i = 1:size(moments2_sorted)[1]
103103
if i % factor == 1
104104
write(f, "$(moments2_sorted[i,1])\t$(sum(moments2_sorted[1:i,2]) ./ length(ds))\n")
105105
end
106106
end
107-
end
107+
end
108108

109109

110110

@@ -116,8 +116,8 @@ times = unique(ds.file[:timestamp])
116116

117117
qnt = collect(1:length(times))
118118

119-
for i=1:length(times)
120-
repeats = length(find(r->r==times[i], ds.file[:timestamp]))
119+
for i = 1:length(times)
120+
repeats = length(find(r->r == times[i], ds.file[:timestamp]))
121121
if repeats > 1
122122
qnt[i] = qnt[i] + repeats - 1
123123
end
@@ -131,33 +131,33 @@ ds = datasets[1]()
131131
sort!(ds.file, cols = :timestamp)
132132
ds.file[:timestamp] = ds.file[:timestamp] .- ds.file[:timestamp][1]
133133

134-
moments = Dict{Int, Array{Int}}()
134+
moments = Dict{Int,Array{Int}}()
135135

136136
times = unique(ds.file[:timestamp])
137137

138-
for i=1:length(times)
138+
for i = 1:length(times)
139139
moments[times[i]] = Array{Int}(0)
140140
end
141141

142-
for (u,v,r,t) in ds
142+
for (u, v, r, t) in ds
143143
push!(moments[t], u)
144144
end
145145

146-
moments2 = Array{Tuple{Int, Int, Int}}(0)
146+
moments2 = Array{Tuple{Int,Int,Int}}(0)
147147

148-
for i=1:length(times)
148+
for i = 1:length(times)
149149
users = moments[times[i]]
150-
for j=1:length(unique(users))
151-
push!(moments2, (times[i], users[j], length(find(r->r==users[j], users))))
152-
println(length(find(r->r==users[j], users)))
150+
for j = 1:length(unique(users))
151+
push!(moments2, (times[i], users[j], length(find(r->r == users[j], users))))
152+
println(length(find(r->r == users[j], users)))
153153
end
154154
end
155155

156156
x = Array{Int}(length(moments2))
157157
y = Array{Int}(length(moments2))
158158
z = Array{Int}(length(moments2))
159159

160-
for i=1:length(moments2)
160+
for i = 1:length(moments2)
161161
x[i] = moments2[i][1]
162162
y[i] = moments2[i][2]
163163
z[i] = moments2[i][3]
@@ -169,11 +169,11 @@ moments = hcat(ds.file[:user], ds.file[:timestamp])
169169
moments[:,2] = moments[:,2] .- moments[1,2]
170170

171171
select = find(r->r in [1:10...], y)
172-
select = (length(x)-1000):length(x)
172+
select = (length(x) - 1000):length(x)
173173
###
174-
scatter(x[select], y[select], s = z[select].*25, alpha = 0.5)
174+
scatter(x[select], y[select], s = z[select] .* 25, alpha = 0.5)
175175

176-
for i=1:10
176+
for i = 1:10
177177
plot(x[select], repeat([i], inner = length(x[select])))
178178
end
179179

0 commit comments

Comments
 (0)