def test_normalization(self):
"""Test z normalizacją."""
prefs_norm_test = {
'10': {u'Beirut': 1.0}, '13': {u'Evanescence': 1.0},
'15': {u'le bowl': 1.0}, '14': {u'The Montesas': 1.0},
'17': {u'Rihanna': 0.1448, u'The Veronicas': 1.0, u'Sia': 0.1886,
u'Britney Spears': 0.1324},
'16': {u'Lena': 1.0},
'18': {u'Eric Clapton': 0.3576, u'The White Stripes': 1.0},
'22': {u'Corpus Christi': 1.0}, '21': {u'Dima Bilan': 1.0},
'2': {u'Röyksopp': 1.0, u'Japan': 0.3655},
'4': {u'אביתר בנאי': 1.0},
'7': {u'Danity Kane': 1.0},
'6': {u'Jill Scott': 0.85, u'Lokua Kanza': 1.0, u'Dru Hill': 0.8},
'9': {u'Katatonia': 1.0},
'20': {u'The Crystal Method': 0.3474, u'Salem': 1.0},
'8': {u'Milow': 0.6525, u'Lily Allen': 1.0}}
dat_file = codecs.open("lastfm.test.dat", "r", "utf-8")
prefs_norm = load_data(dat_file, normalize=True)
dat_file.close()
self.assertTrue("userID" not in prefs_norm)
self.assertEqual(prefs_norm_test, self.round_dict_values(prefs_norm))
def test_no_normalization(self):
"""Test bez normalizacji."""
prefs_test = {
'10': {u'Beirut': 1137.0}, '13': {u'Evanescence': 2.0},
'15': {u'le bowl': 835.0}, '14': {u'The Montesas': 602.0},
'17': {u'Rihanna': 129.0, u'The Veronicas': 891.0, u'Sia': 168.0,
u'Britney Spears': 118.0},
'16': {u'Lena': 264.0},
'18': {u'Eric Clapton': 236.0, u'The White Stripes': 660.0},
'22': {u'Corpus Christi': 1284.0},
'20': {u'The Crystal Method': 99.0, u'Salem': 285.0},
'2': {u'Röyksopp': 3644.0, u'Japan': 1332},
'4': {u'אביתר בנאי': 384.0},
'7': {u'Danity Kane': 1027.0},
'6': {u'Jill Scott': 17.0,
u'Lokua Kanza': 20.0,
u'Dru Hill': 16.0},
'9': {u'Katatonia': 1113.0},
'8': {u'Milow': 400.0, u'Lily Allen': 613.0},
'21': {u'Dima Bilan': 180.0}}
dat_file = codecs.open("lastfm.test.dat", "r", "utf-8")
prefs = load_data(dat_file, normalize=False)
dat_file.close()
self.assertTrue("userID" not in prefs)
self.assertEqual(prefs_test, self.round_dict_values(prefs))
"""Makes recommendations"""
from TaskD01 import load_data
from TaskD03 import sim_pearson
from TaskD04 import get_recommendations
if __name__ == '__main__':
PREFS = load_data(open("lastfm.dat", 'r'), normalize=True)
MY_PREFS = {
"Silverchair": 1,
"Adele": 0.8,
"Britney Spears": -0.5,
"Oasis": 0.7,
"Nirvana": 0.5,
"Metallica": -0.4,
"Depeche Mode": -0.3
}
for i, (n, r) in enumerate(get_recommendations(PREFS,
MY_PREFS, 3, sim_pearson)[:10]):
print "%d\t%s\t%.3f" % (i+1, n, r)
"""Makes recommendations"""
from TaskD01 import load_data
from TaskD03 import sim_pearson
from TaskD04 import get_recommendations
if __name__ == '__main__':
PREFS = load_data(open("lastfm.dat", 'r'), normalize=True)
MY_PREFS = {
"Silverchair": 1,
"Adele": 0.8,
"Britney Spears": -0.5,
"Oasis": 0.7,
"Nirvana": 0.5,
"Metallica": -0.4,
"Depeche Mode": -0.3
}
for i, (n, r) in enumerate(
get_recommendations(PREFS, MY_PREFS, 3, sim_pearson)[:10]):
print "%d\t%s\t%.3f" % (i + 1, n, r)
def test_no_normalization(self):
"""Test bez normalizacji."""
prefs_test = {
'10': {
u'Beirut': 1137.0
},
'13': {
u'Evanescence': 2.0
},
'15': {
u'le bowl': 835.0
},
'14': {
u'The Montesas': 602.0
},
'17': {
u'Rihanna': 129.0,
u'The Veronicas': 891.0,
u'Sia': 168.0,
u'Britney Spears': 118.0
},
'16': {
u'Lena': 264.0
},
'18': {
u'Eric Clapton': 236.0,
u'The White Stripes': 660.0
},
'22': {
u'Corpus Christi': 1284.0
},
'20': {
u'The Crystal Method': 99.0,
u'Salem': 285.0
},
'2': {
u'Röyksopp': 3644.0,
u'Japan': 1332
},
'4': {
u'אביתר בנאי': 384.0
},
'7': {
u'Danity Kane': 1027.0
},
'6': {
u'Jill Scott': 17.0,
u'Lokua Kanza': 20.0,
u'Dru Hill': 16.0
},
'9': {
u'Katatonia': 1113.0
},
'8': {
u'Milow': 400.0,
u'Lily Allen': 613.0
},
'21': {
u'Dima Bilan': 180.0
}
}
dat_file = codecs.open("lastfm.test.dat", "r", "utf-8")
prefs = load_data(dat_file, normalize=False)
dat_file.close()
self.assertTrue("userID" not in prefs)
self.assertEqual(prefs_test, self.round_dict_values(prefs))
def test_normalization(self):
"""Test z normalizacją."""
prefs_norm_test = {
'10': {
u'Beirut': 1.0
},
'13': {
u'Evanescence': 1.0
},
'15': {
u'le bowl': 1.0
},
'14': {
u'The Montesas': 1.0
},
'17': {
u'Rihanna': 0.1448,
u'The Veronicas': 1.0,
u'Sia': 0.1886,
u'Britney Spears': 0.1324
},
'16': {
u'Lena': 1.0
},
'18': {
u'Eric Clapton': 0.3576,
u'The White Stripes': 1.0
},
'22': {
u'Corpus Christi': 1.0
},
'21': {
u'Dima Bilan': 1.0
},
'2': {
u'Röyksopp': 1.0,
u'Japan': 0.3655
},
'4': {
u'אביתר בנאי': 1.0
},
'7': {
u'Danity Kane': 1.0
},
'6': {
u'Jill Scott': 0.85,
u'Lokua Kanza': 1.0,
u'Dru Hill': 0.8
},
'9': {
u'Katatonia': 1.0
},
'20': {
u'The Crystal Method': 0.3474,
u'Salem': 1.0
},
'8': {
u'Milow': 0.6525,
u'Lily Allen': 1.0
}
}
dat_file = codecs.open("lastfm.test.dat", "r", "utf-8")
prefs_norm = load_data(dat_file, normalize=True)
dat_file.close()
self.assertTrue("userID" not in prefs_norm)
self.assertEqual(prefs_norm_test, self.round_dict_values(prefs_norm))
