def setUp(self):
# initialize the logger
job_logger = logging.getLogger('jobs')
job_logger.setLevel(logging.WARNING)
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
ch = logging.StreamHandler(sys.stdout)
ch.setFormatter(formatter)
job_logger.addHandler(ch)
cfg = config.load_config('config.cfg')
self.server = server.Server(cfg)
def setUp(self):
cfg = config.load_config('config.cfg')
self.server = server.Server(cfg)
import numpy
import redisml.server.server as server
import redisml.server.configuration as config
def gnnmf(v, w, h):
max_iteration = 10
for i in range(0, max_iteration):
# Numpy code:
# h = h*(w.transpose().dot(h)/w.transpose().dot(w).dot(h))
# w = w*(v.dot(h.transpose())/w.dot(h).dot(h.transpose()))
h = h.cw_multiply(w.multiply(h, transpose_self=True).cw_divide(w.multiply(w, transpose_self=True).multiply(h)))
w = w.cw_multiply(v.multiply(h, transpose_m=True).cw_divide(w.multiply(h).multiply(h, transpose_m=True)))
cfg = config.load_config('config.cfg')
s = server.Server(cfg)
# Here we generate random matrices
# Use Server.matrix_from_name to load an existing matrix from redis
v_ = numpy.random.rand(1024,1024)
v = s.matrix_from_numpy(v_)
w_ = numpy.random.rand(1024,1024)
w = s.matrix_from_numpy(w_)
h_ = numpy.random.rand(1024,1024)
h = s.matrix_from_numpy(h_)
gnnmf(v, w, h)
print h
import numpy
import redisml.server.server as server
import redisml.server.configuration as config
def kmeans(data, centers, k):
# iterate k-means
for i in range(0, k):
dist = data.k_means_distance(centers)
centers = data.k_means_recalc(dist)
# Create a server
cfg = config.load_config("config.cfg")
s = server.Server(cfg)
# Generate random data
data_ = numpy.random.rand(1024, 16)
# Take first 10 rows as centers
centers_ = data_[0:10, :]
# Create redisml matrices
data = s.matrix_from_numpy(data_)
centers = s.matrix_from_numpy(centers_)
kmeans(data, centers, 10)
print centers
