# pickle fails for nested functions
def adder(augend):
zero = [0]
def inner(addend):
return addend+augend+zero[0]
return inner
# build from inner function
add_me = adder(5)
# build from lambda functions
squ = lambda x:x**2
# test 'dilled' multiprocessing for inner
print "Evaluate 10 items on 2 proc:"
pool.ncpus = 2
print pool
print pool.map(add_me, range(10))
print ''
# test 'dilled' multiprocessing for lambda
print "Evaluate 10 items on 4 proc:"
pool.ncpus = 4
print pool
print pool.map(squ, range(10))
print ''
# test for lambda, but with threads
print "Evaluate 10 items on 4 threads:"
tpool.nthreads = 4
print tpool
#!/usr/bin/env python
from pathos.multiprocessing import ProcessingPool as Pool
from pathos.multiprocessing import ThreadingPool as TPool
pool = Pool()
tpool = TPool()
def host(id):
import socket
return "Rank: %d -- %s" % (id, socket.gethostname())
print "Evaluate 10 items on 1 proc"
pool.ncpus = 1
res3 = pool.map(host, range(10))
print pool
print '\n'.join(res3)
print ''
print "Evaluate 10 items on 2 proc"
pool.ncpus = 2
res5 = pool.map(host, range(10))
print pool
print '\n'.join(res5)
print ''
print "Evaluate 10 items on ? proc"
pool.ncpus = None
res7 = pool.map(host, range(10))
print pool
print '\n'.join(res7)
print("False Negative: %d" % (sum((pred_fc == False) & (train_label == 1))))
# if we do not have trained classfiers, train
else:
print("boosters do not exist")
print("start finding features")
# contruct all filters
wdw_stride = 2
port = 0.75
filters = GetFilter(wdw_stride, port)
# parallel computing setup
pool = Pool()
pool.ncpus = 16
# train classifiers and dump to file
bstrs = open("./boosters.pkl", 'wb')
tr_time = open("./train_time.pkl", 'wb')
boosters = []
num_booster = 4
pred_fc = (np.ones((2 * n,)) == 1)
for i in range(num_booster):
boosters.append(AFS.AdaFeatureSelection(None, None))
stime = time()
boosters[-1].train(train_integral[pred_fc, :, :], train_label[pred_fc], filters, pool)
pred = boosters[-1].predict(train_integral[pred_fc, :, :])
pred_fc[pred_fc == True] = (pred == 1)
duration = (time() - stime) / 60
print("classifier %i: training time %.3f min" % (i+1, duration))
return data_channel_default
# Determine dimensions of 2D grid
shape2d = dat3d.shape[1:]
# Create tuples containing all the combinations of indices
indexlist = list(it.product(range(shape2d[0]), range(shape2d[1])))
# Dictionary to save the indices
wrapdictlist = []
for indices in indexlist:
d = {}
d['indices'] = indices
wrapdictlist.append(d)
pool = ProcessingPool()
pool.ncpus = 6
output_list = pool.map(wrapperfunction, wrapdictlist)
# Now create arrays for data storage
qc_status_int = np.zeros(shape2d, dtype='float')
mk_pvalue = np.zeros(shape2d, dtype='float')
algorithm_succes = np.zeros(shape2d, dtype='int')
# Now start unpacking the results
for d in output_list:
i, j = d['indices']
qc_status_int[i, j] = d['qc_status_int']
mk_pvalue[i, j] = d['mk_pvalue']
algorithm_succes[i, j] = d['algorithm_succes']
plt.figure(figsize=(20, 15))
def inner(addend):
return addend + augend + zero[0]
return inner
# build from inner function
add_me = adder(5)
# build from lambda functions
squ = lambda x: x**2
# test 'dilled' multiprocessing for inner
print "Evaluate 10 items on 2 proc:"
pool.ncpus = 2
print pool
print pool.map(add_me, range(10))
print ''
# test 'dilled' multiprocessing for lambda
print "Evaluate 10 items on 4 proc:"
pool.ncpus = 4
print pool
print pool.map(squ, range(10))
print ''
# test for lambda, but with threads
print "Evaluate 10 items on 4 threads:"
tpool.nthreads = 4
print tpool
# License: 3-clause BSD. The full license text is available at:
# - http://trac.mystic.cacr.caltech.edu/project/pathos/browser/pathos/LICENSE
from pathos.multiprocessing import ProcessingPool as Pool
from pathos.multiprocessing import ThreadingPool as TPool
pool = Pool()
tpool = TPool()
def host(id):
import socket
return "Rank: %d -- %s" % (id, socket.gethostname())
print "Evaluate 10 items on 1 proc"
pool.ncpus = 1
res3 = pool.map(host, range(10))
print pool
print '\n'.join(res3)
print ''
print "Evaluate 10 items on 2 proc"
pool.ncpus = 2
res5 = pool.map(host, range(10))
print pool
print '\n'.join(res5)
print ''
print "Evaluate 10 items on ? proc"
pool.ncpus = None
res7 = pool.map(host, range(10))
