MovieLensVisualization/hw5method.py at master · jaceyca/MovieLensVisualization · GitHub

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
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
import numpy as np
import matplotlib.pyplot as plt
from prob2utils import train_model, get_err, get_err_advanced
from visualization import get_rating_freq, bar_plot, fancy_plot, matrix_factorization_visualization
from numpy.linalg import svd
import pylab as pl
import codecs
import operator
from collections import Counter


def projectUV(U, V):
    '''
    This function projects U and V in to a 2D space so we can create visualizations.

    Input:
        U: The U matrix from SVD
        V: The V matrix from SVD

    Output:
        newU: The 2D version of U
        newV: The 2D version of V
    '''

    A, S, B = svd(V)
    A = np.array(A)
    A = A[:, [0, 1]]
    newU = np.dot(np.transpose(A), np.transpose(U))
    newV = np.dot(np.transpose(A), V)

    return newU, newV


def main():
    Y_train = np.loadtxt('./data/traintest.txt').astype(int)
    Y_test = np.loadtxt('./data/test.txt').astype(int)
    data = np.loadtxt('./data/data.txt').astype(int)
    movie_file = codecs.open('./data/movies.txt', mode='r', encoding='windows-1252')
    movie_names = {}
    genres = {}
    for line in movie_file:
        movie_info = line.split()
        movie_names[int(movie_info[0])] = " ".join(movie_info[1:-19])
        genres[int(movie_info[0])] = list(map(int, movie_info[-19:]))


    M = max(max(Y_train[:,0]), max(Y_test[:,0])).astype(int) # users
    N = max(max(Y_train[:,1]), max(Y_test[:,1])).astype(int) # movies
    Ks = [20]

    # Ein and Eout for different regs have been recorded
    # 2*10**-1 gives us an eout of 0.5 but it has meaningful graphs?
    regs = [2*10**-1]
    eta = 0.03 # learning rate
    E_ins = []
    E_outs = []

    # Use to compute Ein and Eout
    for reg in regs:
        E_ins_for_lambda = []
        E_outs_for_lambda = []

        for k in Ks:
            print("Training model with M = %s, N = %s, k = %s, eta = %s, reg = %s"%(M, N, k, eta, reg))
            # U,V, e_in = train_model(M, N, k, eta, reg, Y_train, mode='basic')
            # E_ins_for_lambda.append(e_in)
            # eout = get_err(U, V, Y_test)

            U,V, e_in, aVec, bVec, mu = train_model(M, N, k, eta, reg, Y_train, mode='advanced')
            E_ins_for_lambda.append(e_in)
            eout = get_err_advanced(U, V, Y_test, mu, aVec, bVec, reg)

            E_outs_for_lambda.append(eout)

        E_ins.append(E_ins_for_lambda)
        E_outs.append(E_outs_for_lambda)

    # basic gives Ein = 0.3002 and Eout = 0.4495
    # advanced gives Ein = 0.4254 and Eout = 0.5097


    newU, newV = projectUV(U, V)
    return newU, newV

'''
    preds = np.dot(np.transpose(newU), newV)

    for index, row in enumerate(preds):
        row[:] = row + aVec[index] + mu
    for index, column in enumerate(np.transpose(preds)):
        column[:] = column + bVec[index]
    intPreds = np.around(preds).astype(int)
    flatPreds = list(np.ndarray.flatten(intPreds))

    print('Errors:')
    print(E_ins)
    print(E_outs)

    r1 = flatPreds.count(1)
    r2 = flatPreds.count(2)
    r3 = flatPreds.count(3)
    r4 = flatPreds.count(4)
    r5 = flatPreds.count(5)
    f1 = r1/(r1+r2+r3+r4+r5)
    f2 = r2/(r1+r2+r3+r4+r5)
    f3 = r3/(r1+r2+r3+r4+r5)
    f4 = r4/(r1+r2+r3+r4+r5)
    f5 = r5/(r1+r2+r3+r4+r5)

    rating_freq = [f1, f2, f3, f4, f5]
    # bar_plot(rating_freq, "Advanced Ratings of All Predicted Movies")


    # Plotting for actual things

    rand_movie_ids = np.random.randint(1, 1682, 10)
    rand_movie_names = []
    for movieID in rand_movie_ids:
        rand_movie_names.append(movie_names[movieID])
    matrix_factorization_visualization(newV, rand_movie_ids, rand_movie_names, "Advanced Predictions of Random Movies")

    pop_movie_ids = [50, 258, 100, 181, 294, 286, 288, 1, 300, 121]
    pop_movie_names = ['Star Wars (1977)', 'Contact (1997)', 'Fargo (1996)', 'Return of the Jedi (1983)',
    'Liar Liar (1997)', '"English Patient, The (1996)"', 'Scream (1996)', 'Toy Story (1995)',
    'Air Force One (1997)', 'Independence Day (ID4) (1996)']

    matrix_factorization_visualization(newV, pop_movie_ids, pop_movie_names, "Advanced Predictions of Popular Movies")

    best_movie_ids = [1189, 1500, 814, 1536, 1293, 1599, 1653, 1467, 1122, 1201]
    best_movie_names = ['Prefontaine (1997)', 'Santa with Muscles (1996)', '"Great Day in Harlem, A (1994)"',
    'Aiqing wansui (1994)', 'Star Kid (1997)', "Someone Else's America (1995)",
    'Entertaining Angels: The Dorothy Day Story (1996)', '"Saint of Fort Washington, The (1993)"',
    'They Made Me a Criminal (1939)', 'Marlene Dietrich: Shadow and Light (1996)']

    matrix_factorization_visualization(newV, best_movie_ids, best_movie_names, "Advanced Predictions of Best Movies")

    action_movie_ids = [2, 21, 24, 29, 35, 50, 62, 78, 82, 97, 101, 110, 112, 117, 118, 132, 140, 141, 142,
    148, 151, 164, 172, 173, 174, 181, 184, 201, 206, 210, 222, 227, 228, 229, 230, 231, 247, 252, 254, 257,
    266, 271, 304, 331, 355, 358, 373, 380, 385, 389, 398, 399, 403, 405, 431, 449, 450, 457, 463, 465, 472,
    491, 495, 498, 500, 511, 519, 520, 526, 530, 541, 554, 560, 562, 566, 576, 601, 622, 636, 655, 679, 680,
    720, 755, 768, 769, 802, 812, 820, 826, 827, 828, 829, 831, 862, 877, 890, 897, 919, 924, 930, 947, 951,
    982, 993, 1013, 1016, 1031, 1033, 1034, 1058, 1060, 1076, 1091, 1105, 1116, 1126, 1133, 1239, 1279, 1293,
    1314, 1383, 1411, 1450, 1469, 1480, 1503, 1515, 1523, 1531, 1540, 1555, 1608, 1615]

    matrix_factorization_visualization(newV, action_movie_ids, None, "Advanced Predictions of Adventure Movies")

    documentary_movie_ids = [32, 48, 75, 115, 119, 320, 360, 634, 644, 645, 677, 701, 757, 766, 811, 813, 814,
    847, 850, 857, 884, 954, 973, 1022, 1065, 1084, 1128, 1130, 1141, 1142, 1184, 1201, 1232, 1294, 1307, 1318,
    1331, 1363, 1366, 1378, 1482, 1497, 1547, 1561, 1562, 1585, 1594, 1629, 1641, 1649]

    matrix_factorization_visualization(newV, documentary_movie_ids, None, "Advanced Predictions of Documentaries")

    war_movie_ids = [10, 22, 31, 50, 51, 69, 80, 110, 121, 133, 157, 172, 176, 180, 181, 188, 190, 199, 205,
    211, 214, 235, 241, 245, 271, 286, 318, 326, 426, 430, 471, 47, 4, 483, 498, 502, 511, 515, 520, 521, 528,
    549, 593, 601, 631, 641, 647, 651, 687, 690, 719, 744, 803, 879, 891, 935, 944, 971, 1065, 1124, 1152, 1176,
    1185, 1204, 1357, 1423, 1485, 1501, 1529, 1574, 1632, 1663]

    matrix_factorization_visualization(newV, war_movie_ids, None, "Advanced Predictions of War Movies")
'''
if __name__ == "__main__":
    main()