def __init__(self,
matrix,
max_neighbours=20,
lshashes=[RandomBinaryProjections("rbp", 10)],
vector_filters=[UniqueFilter()],
distance=Pearson()):
if not isinstance(lshashes, list):
raise TypeError("'lshashes' must be an instance of 'list'")
if not isinstance(vector_filters, list):
raise TypeError("'vector_filters' must be an instance of 'list'")
self.underlying = Engine(len(matrix[0]),
lshashes=lshashes,
vector_filters=vector_filters +
[NearestFilter(max_neighbours)],
distance=distance)
for vector in matrix:
self.underlying.store_vector(vector)
def main(args):
Pearson(args[1:]).global_()
import sys
from pearson import Pearson
def main(args):
Pearson(args[1:]).global_()
if __name__ == '__main__':
Pearson(sys.argv[1:]).global_()
def main():
if len(sys.argv) == 1:
K = 200
else:
K = int(sys.argv[1])
pearObj = Pearson("test5.txt", K)
pearObj.compute()
print("test5.txt processed. result5.txt produced")
pearObj = Pearson("test10.txt", K)
pearObj.compute()
print("test10.txt processed. result10.txt produced")
pearObj = Pearson("test20.txt", K)
pearObj.compute()
print("test20.txt processed. result20.txt produced")
def train(users, algorithm):
if algorithm == 'pearson':
model = Pearson()
model.train(users)
return model
elif algorithm == 'spearman':
model = Spearman()
model.train(users)
return model
elif algorithm == 'mean_squared':
model = Meansquared()
model.train(users)
return model
elif algorithm == "cosine":
model = Cosine()
model.train(users)
return model
else:
print("No model found given algorithm " + algorithm)
sys.exit(-1)
def main(args):
Pearson(args[1:]).local_()
import sys
from pearson import Pearson
def main(args):
Pearson(args[1:]).local_()
if __name__ == "__main__":
Pearson(sys.argv[1:]).local_()
def main():
pearObj = Pearson("test5.txt", 200)
pearObj.compute(False, True)
print("test5.txt processed. result5.txt produced")
pearObj = Pearson("test10.txt", 200)
pearObj.compute(False, True)
print("test10.txt processed. result10.txt produced")
pearObj = Pearson("test20.txt", 200)
pearObj.compute(False, True)
print("test20.txt processed. result20.txt produced")