def test_unaryFunctions(self):
for func,inOutData in TEST_CASES_UNARY.items():
print('testing univar_polyops.%s with numpy: ' % func.__name__, end='')
for output,input in inOutData:
self.assertTrue(na_eq(na(output), func(na(input))), 'testing %s = %s(%s)' % (output,func.__name__,input))
print('.', end='')
print()
def add_ids(json_list):
ids = list(
map(
bson.Int64,
list(
na([get_max_id()] * len(json_list)) +
na(list(range(1,
len(json_list) + 1))))))
return list(map(add_id_, list(zip(ids, json_list))))
def getFitForN(N):
a = []
for i in range(1, N + 1):
row = []
for j in range(N - 1, -1, -1):
row.append(pow(i, j))
a.append(row)
a = na(a)
b = na([sequence(i) for i in range(1, N + 1)])
coeffs = solve(a, b)
fit = pv(coeffs, N + 1)
return fit
def get_duplicates(update=False):
dups, _ = get_duplicates_for_range(1900, 2017)
duplicates = na(list(map(lambda x: x[0], dups)))
if not update:
return duplicates
with open(duplicate_fname, 'wb') as f:
np.save(f, duplicates)
return duplicates
def save_processed_pmids():
existing = get_existing_pmids()
downloaded = get_downloaded_pmids()
result = existing.union(downloaded)
arr_to_save = na(list(result))
with open(processed_fname, 'wb') as f:
np.save(f, arr_to_save)
return result
def get_data_from_dcm(self, idx):
dcms = []
parent_path = self.ids[idx]
for file in os.listdir(parent_path):
if not file.endswith('dcm'):
continue
image, dcm_data = imread(opjoin(parent_path, file))
if not has_slice_location(dcm_data):
continue
dcms.append((image, dcm_data))
dcms.sort(key=lambda dcm: dcm[1].SliceLocation)
nodules = parseXML(parent_path)
id2roi = create_map_from_nodules(nodules)
imgs = na([dcm[0] for dcm in dcms])
imgs = imgs[np.newaxis, :]
bbox = resolve_bbox(na(dcms), id2roi)
return imgs, bbox
def basiccube():
dir = na([[1,0,0]])
k = 15
maxvol = 0.02
g = dg3d.impedance(dir,k)
gg = dg3d.planeWaves(dir,k)[0]
s = dg3d.solver(dg3d.cubemesh(maxvol), 2, 12, k).solve(g)
v = dg3d.Visualiser(s,gg)
def numpysum():
import time
n = 1000
A = numpy.mat(numpy.reshape(range(n*n), (n,n)), dtype=numpy.float64)
t = []
t.append(time.time())
B1 = A + A +A + A
t.append(time.time())
A2 = [A,A,A,A]
t.append(time.time())
B2 = numpy.sum(A2, axis=0)
t.append(time.time())
l = len(A2)
B3 = numpy.reshape(numpy.dot(numpy.ones(l), numpy.reshape(A2, (l, -1))), (n,n))
t.append(time.time())
op = lambda a,b:a+b
B4 = reduce(op, A2)
t.append(time.time())
print na(t[1:]) - na(t[:-1])
def save_downloaded_pmids():
downloaded_pmids = set(get_downloaded_pmids_for_range(1900, 2017))
arr_to_save = na(list(downloaded_pmids))
with open(downloaded_fname, 'wb') as f:
np.save(f, arr_to_save)
return downloaded_pmids
def getU(t):
if t < 5:
return u
elif 20 < t < 22:
return -u / 2
else:
return 0
deltaT = 1
D = 100
M = 1e4 # 1 ton
# system state
F = na([[1, deltaT], [0, M / (M + deltaT * D)]])
# input matrix
G = nvect(.5 * deltaT * deltaT, deltaT)
# observation matrix
H = na([[1, 0]])
# process noise covariance
Q = np.eye(2) * .1
# measurement noise covar.
R = na([100])
# true init state
x_0 = nvect(0, 0)
# assumed init state
x_0_hat = nvect(50, 0)
# assumed init state err covar matrix
P = na([[1000, 0], [0, 1]])
from numpy.random import randn
import numpy as np
import matplotlib.pyplot as plt
def nvect(*argv):
return np.array([argv]).T
cmap = plt.get_cmap("tab20c")
""" process model """
u = 10. #m/s^2
deltaT = 1e-2
# system state
F = na([[1, deltaT], [0, 1]])
# input matrix
G = nvect(-.5 * deltaT * deltaT, -deltaT)
# observation matrix
H = na([[1, 0]])
""" uncertainty """
n = 2 # number of states
q = 0.5 # std of process
r = 2 # std of measurement
# process noise covariance
Q = q**2 * np.eye(n) # covariance of process
#measurement noise covar.
R = r**2 # covariance of measurement
#Q=np.zeros(2)
#R=na([4])
A2 = [A,A,A,A]
t.append(time.time())
B2 = numpy.sum(A2, axis=0)
t.append(time.time())
l = len(A2)
B3 = numpy.reshape(numpy.dot(numpy.ones(l), numpy.reshape(A2, (l, -1))), (n,n))
t.append(time.time())
op = lambda a,b:a+b
B4 = reduce(op, A2)
t.append(time.time())
print na(t[1:]) - na(t[:-1])
if __name__ == '__main__':
print "hello"
dir = na([[1,0,0]])
k = 20
maxvol = 0.04
print k, maxvol
g = dg3d.impedance(dir,k)
gg = dg3d.planeWaves(dir,k)[0]
s = dg3d.solver(dg3d.cubemesh(maxvol), 4, 10, k)
s.solve(g)
v = dg3d.Visualiser(s,gg)
v.showuaveragereal()
sys.exit(0)
- Code By Michael Sherif Naguib
- license: MIT open source
- Date: 12/20/18
- @University of Tulsa
- Description: for storing constants of functions...
'''
# ========================== Predefined Transformations =======================
# Predefined Transformations: [MATRIX as [[A,B],[C,D]], SHIFTS as [X,Y], ROTATION ,PROBABILITY]
from numpy import array as na # NOT NP... NOTE: Numpy Array as na
from math import *
constants = {}
#Barnsley Fern
constants["bfern"] = [
[na([[0, 0], [0, 0.16]]), na([0, 0]), 0, 0.01],
[na([[0.85, 0.04], [0 - 0.04, 0.85]]),
na([0, 1.6]), 0, 0.85],
[na([[0.2, 0 - 0.26], [0.23, 0.22]]),
na([0, 1.6]), 0, 0.07],
[na([[0 - 0.15, 0.28], [0.26, 0.24]]),
na([0, 0.44]), 0, 0.07]
]
#Rose(like)
constants["rose"] = [
[na([[0, 0], [0, 0.16]]), na([0, 0]), 0, 0.01],
[na([[0.85, 0.04], [0 - 0.04, 0.85]]),
na([0, 1.6]), 45, 0.85],
[na([[0.2, 0 - 0.26], [0.23, 0.22]]),
na([0, 1.6]), 3, 0.10],
