def predict_rating(probID, userID, MIN_RATING, MAX_RATING, SVDNeighbourhood=False): if SVDNeighbourhood: svd2 = SVDNeighbourhood() svd2.load_model(os.path.join(utils.get_add_dir(), 'ratings_neigh')) else: svd2 = SVD() svd2.load_model(os.path.join(utils.get_add_dir(), 'ratings')) return svd2.predict(probID, userID, MIN_RATING, MAX_RATING)
class KNNPlusSVD_lib:
def __init__(self, filename, K):
self.svd = SVDNeighbourhood()
self.K = K
self.svd.load_data(filename , sep=' ', format={'col':0, 'row':1, 'value':2, 'ids': int})
def predict(self, userId, itemId):
self.svd.compute(self.K, min_values=5, pre_normalize='all' , mean_center=True, post_normalize=None)
r = self.svd.predict(11, 33, weighted=True, MIN_VALUE=1.0, MAX_VALUE=5.0)
return r
# recommender = Recommender()
# recommender.load_web_data('dataset',
# [{'Запах женщины': 9, 'The Usual Suspects': 8, 'The Departed': 8,
# 'Тутси': 7, 'Выпускник': 10, 'Залечь на дно в Брюгге': 4, 'Евротур': 7,
# 'Goodfellas': 6, 'Донни Браско': 8, 'Амели': 3, 'Идиократия': 7}],
# 100, 0, 10, 10)
# recommender.load_local_data('dataset', K=100, min_values=0)
# m = recommender.matrix.get_rating_matrix()
#
# m1 = recommender.get_predictions_for_all_users()
from recsys.algorithm.factorize import SVDNeighbourhood
svd = SVDNeighbourhood()
svd.load_data('test_dataset',
sep=' ',
format={
'col': 1,
'row': 0,
'value': 2,
'ids': int
})
svd.compute(100, 0)
print svd.predict(108, 698)
# svd.load_data(filename=sys.argv[1], sep='::', format={'col':0, 'row':1, 'value':2, 'ids': int})
# K=100
# svd.compute(k=K, min_values=5, pre_normalize=None, mean_center=True, post_normalize=True)
# Evaluate
rmse_svd = RMSE()
mae_svd = MAE()
rmse_svd_neig = RMSE()
mae_svd_neig = MAE()
i = 1
total = len(test.get())
print "Total Test ratings: %s" % total
for rating, item_id, user_id in test:
try:
pred_rating_svd = svd.predict(item_id, user_id)
rmse_svd.add(rating, pred_rating_svd)
mae_svd.add(rating, pred_rating_svd)
pred_rating_svd_neig = svd_neig.predict(item_id, user_id) # Koren & co.
if pred_rating_svd_neig is not nan:
rmse_svd_neig.add(rating, pred_rating_svd_neig)
mae_svd_neig.add(rating, pred_rating_svd_neig)
print "\rProcessed test rating %d" % i,
sys.stdout.flush()
i += 1
except KeyError:
continue
rmse_svd_all.append(rmse_svd.compute())
mae_svd_all.append(mae_svd.compute())
rmse_svd_neig_all.append(rmse_svd_neig.compute())
mae_svd_neig_all.append(mae_svd_neig.compute())
sum_value = 0.0
for i in range(1,11):
#Train & Test data
train, test = data.split_train_test(percent=PERCENT_TRAIN)
K=j
svd = SVDNeighbourhood()
svd.set_data(train)
svd.compute(k=K, min_values=5, pre_normalize=None, mean_center=True, post_normalize=True)
#Evaluation using prediction-based metrics
rmse = RMSE()
mae = MAE()
for rating, item_id, user_id in test.get():
try:
pred_rating = svd.predict(item_id, user_id)
rmse.add(rating, pred_rating)
mae.add(rating, pred_rating)
except KeyError:
continue
print 'RMSE=%s' % rmse.compute()
sum_value = sum_value + rmse.compute()
print '-------'
print 'the k value is %s' %j
print 'Final RMSE=%s' % sum_value
print '-------'
sum_value = sum_value/10
list.append(sum_value)
print 'i value is'
if __name__ == "__main__":
# recommender = Recommender()
# recommender.load_web_data('dataset',
# [{'Запах женщины': 9, 'The Usual Suspects': 8, 'The Departed': 8,
# 'Тутси': 7, 'Выпускник': 10, 'Залечь на дно в Брюгге': 4, 'Евротур': 7,
# 'Goodfellas': 6, 'Донни Браско': 8, 'Амели': 3, 'Идиократия': 7}],
# 100, 0, 10, 10)
# recommender.load_local_data('dataset', K=100, min_values=0)
# m = recommender.matrix.get_rating_matrix()
#
# m1 = recommender.get_predictions_for_all_users()
from recsys.algorithm.factorize import SVDNeighbourhood
svd = SVDNeighbourhood()
svd.load_data('test_dataset', sep=' ', format={'col': 1, 'row': 0, 'value': 2, 'ids': int})
svd.compute(100, 0)
print svd.predict(108, 698)
# svd.load_data(filename=sys.argv[1], sep='::', format={'col':0, 'row':1, 'value':2, 'ids': int})
# K=100
# svd.compute(k=K, min_values=5, pre_normalize=None, mean_center=True, post_normalize=True)
# Evaluate
rmse_svd = RMSE()
mae_svd = MAE()
rmse_svd_neig = RMSE()
mae_svd_neig = MAE()
i = 1
total = len(test.get())
print 'Total Test ratings: %s' % total
for rating, item_id, user_id in test:
try:
pred_rating_svd = svd.predict(item_id, user_id)
rmse_svd.add(rating, pred_rating_svd)
mae_svd.add(rating, pred_rating_svd)
pred_rating_svd_neig = svd_neig.predict(item_id,
user_id) #Koren & co.
if pred_rating_svd_neig is not nan:
rmse_svd_neig.add(rating, pred_rating_svd_neig)
mae_svd_neig.add(rating, pred_rating_svd_neig)
print "\rProcessed test rating %d" % i,
sys.stdout.flush()
i += 1
except KeyError:
continue
rmse_svd_all.append(rmse_svd.compute())
mae_svd_all.append(mae_svd.compute())
rmse_svd_neig_all.append(rmse_svd_neig.compute())
mae_svd_neig_all.append(mae_svd_neig.compute())
__author__ = 'ponomarevandrew'
from recsys.algorithm.factorize import SVDNeighbourhood
svd = SVDNeighbourhood()
svd.load_data(filename='ml-100k/u1.base', sep=' ', format={'col':0, 'row':1, 'value':2, 'ids': int})
K=30
svd.compute(k=K, min_values=5, pre_normalize='all' , mean_center=True, post_normalize=None)
print(svd.predict(11, 33, weighted=True, MIN_VALUE=1.0, MAX_VALUE=5.0))
for j in range(0,300,50):
sum_value = 0.0
for i in range(1,11):
#Create SVD
K= j
train, test = data.split_train_test(percent=PERCENT_TRAIN)
svd = SVDNeighbourhood()
svd.set_data(train)
svd.compute(k=K, min_values=20, pre_normalize=None, mean_center=True, post_normalize=True)
#Evaluation using prediction-based metrics
rmse = RMSE()
mae = MAE()
for rating, item_id, user_id in test.get():
try:
pred_rating = svd.predict(item_id, user_id, weighted=True, MIN_VALUE=0.0, MAX_VALUE=5.0)
rmse.add(rating, pred_rating)
mae.add(rating, pred_rating)
except KeyError:
continue
print 'RMSE=%s' % rmse.compute()
sum_value = sum_value + rmse.compute()
print '-------'
print 'the k value is %s' %j
print 'Final RMSE=%s' % sum_value
print '-------'
sum_value = sum_value/10
list.append(sum_value)
print 'i value is'
for i in range(0,300,50):
