def testsort(self):
bsort = sorting.BubbleSort()
qsort = sorting.QuickSort()
isort = sorting.InsertionSort()
ssort = sorting.SelectionSort()
bsort.array = qsort.array = isort.array = ssort.array = [5, 9, 3, 7, 2, 7, 2, 1, 5]
bsort.sorting()
isort.sorting()
ssort.sorting()
self.assertEqual(bsort.array, [1, 2, 2, 3, 5, 5, 7, 7, 9])
self.assertEqual(qsort.array, [1, 2, 2, 3, 5, 5, 7, 7, 9])
self.assertEqual(isort.array, [1, 2, 2, 3, 5, 5, 7, 7, 9])
self.assertEqual(ssort.array, [1, 2, 2, 3, 5, 5, 7, 7, 9])
array = []
while len(array) <= random.randint(10000, 100000):
array.append(random.randint(0, 100000))
print("Len:", len(array))
bsort = sorting.BubbleSort()
qsort = sorting.QuickSort()
isort = sorting.InsertionSort()
ssort = sorting.SelectionSort()
bsort.array = qsort.array = isort.array = ssort.array = array
bsort.sorting()
isort.sorting()
ssort.sorting()
self.assertEqual(bsort.array, sorted(array))
self.assertEqual(qsort.array, sorted(array))
self.assertEqual(isort.array, sorted(array))
self.assertEqual(ssort.array, sorted(array))
def debug():
lst = []
for i in range(15):
lst.append(random.randint(0, 20))
event_bus = events.EventBus()
sorter = sorting.BubbleSort(sorting.ListState(lst, event_bus))
sorter.sort()
print(lst)
def testBubbleSort(self):
print "\ntest Bubble sort simple case"
m = n
sorter = sorting.BubbleSort(m)
sorter.SetRefColNum(0)
sorter.sort_data()
arr = []
for i in range(length):
arr.append(m.GetVal(i, 0))
self.assertTrue(non_decreasing(arr))
def findC(datafiles, testspecies, bestC, options):
# Interpolate each datafile, generate a matrix from interpolated data
nofiles = len(datafiles)
nocols = len(testspecies) + 1
locs = np.zeros(nocols - 1)
interpdata = np.zeros((nofiles, nocols))
filesmatrix = np.zeros((nofiles, 3))
for ii in range(nofiles): # interpolate for each file
dataobj = iof.ProcFile(datafiles[ii])
if ii == 0:
titles = dataobj.gettitles()
else:
titles1 = dataobj.gettitles()
np.testing.assert_array_equal(
titles1, titles
) # Verify that all data files have the same column headers
dataobj.interpolate(testspecies,
locs,
interpdata[ii, :],
interpval=float(options["StoichMassFrac"][0]),
interpmethod="".join(options["InterpMethod"]))
filesmatrix[:, 0] = interpdata[:, 0]
filesmatrix[:,
2] = interpdata[:,
0] # filesmatrix stores (row1) stoich prog var (row2) file indices (row3) stoich Temps
filesmatrix[:, 1] = range(nofiles)
filesmatC = matrix.Matrix(nofiles, 3)
for i in range(nofiles):
for j in range(3):
filesmatC.SetVal(i, j, filesmatrix[i, j])
# Generate combinations matrix - skip this step if user input says to
skip = options["SkipProgVar"][0]
assert (
(skip == 'yes') | (skip == 'no')
), "must select input <yes> or <no> for <skip progress variable optimization:>"
if skip == 'no':
combosmatrix = np.zeros((nocols, cs.totnumcoms(nocols - 1) + 1))
cs.combination_mat(combosmatrix[1:, 1:])
else:
print "Skipping progress variable optimization, using user input"
options["PlotAllC"][0] = 'no'
combosmatrix = np.zeros((nocols, 2))
for i in range(nocols - 1):
combosmatrix[i + 1, 1] = 1
combosmatrix[0, 0] = 1
# Calculate progress variables
progvars = np.dot(interpdata, combosmatrix)
length = progvars.shape[1]
# Generate progress variable matrix
progVar = matrix.Matrix(nofiles, length)
for i in range(nofiles):
for j in range(length):
progVar.SetVal(i, j, progvars[i, j])
# Sort PROGVARS and FILESMATRIX by temperature
sortmethod = options["sort method"][0]
print "Sorting PROGVARS by temperature using %s sort" % sortmethod
for i in [progVar, filesmatC]:
if sortmethod == 'bubble':
sorter = sorting.BubbleSort(i)
elif sortmethod == 'standard':
sorter = sorting.StandardSort(i)
elif sortmethod == 'brute':
sorter = sorting.BruteSort(i)
else:
raise IOError(
"invalid sorting method (%s) specified, instead use <bubble> sort"
% sortmethod)
sorter.SetRefColNum(0)
sorter.sort_data()
print "Sorting FILESMATRIX by temperature"
# Test monotonicity of PROGVARS
print "Testing monotonicity \n"
monoAry = np.zeros(length, dtype=np.int32)
checker = monocheck.MonoCheck(progVar) # Create MonoCheck object
assert checker.CheckStrictMonoticity(
monoAry, 0) == 0, "CheckStrictMonoticity ran unsuccessfully.\n"
# ^ Check which columns of progVar are strictly increasing or strictly decreasing and store result in monoAry
# Test for maximum slope if multiple monotonic progress variables are returned
checksum = np.sum(monoAry)
if checksum % 3 != 0:
raise RuntimeError(
"Incorrect values in monoAry vector, check monotonicity function.\n"
)
if checksum > 3:
maxslopetest = options["MaxSlopeTest"][0]
print "Testing max slope using %s" % maxslopetest
if maxslopetest == 'linear regression':
maxchecker = maxslope.LinRegression(progVar)
elif maxslopetest == 'end point slope':
maxchecker = maxslope.EndPointSlope(progVar)
else:
raise IOError(
"invalid maximum slope test (%s) specified, instead use <linear regression>"
% maxslopetest)
assert maxchecker.MostMonotonic(
monoAry, 0) == 0, "MostMonotonic ran unsuccessfully.\n"
# ^ Distinguish the best monotonic progress variables
elif checksum == 0:
# Least non-monotonic tests to be implemented in beta version
lnmoption = options["lnmcheck"][0]
print "Finding least non-monotonic progress variable using %s nonmono check" % lnmoption
if lnmoption == 'simple':
lnm_check = leastnonmono.SimpleLNM(progVar)
elif lnmoption == 'advanced':
lnm_check = leastnonmono.AdvancedLNM(progVar)
else:
raise IOError(
"invalid lnm test (%s) specified, instead use <simple> or <advanced>"
% maxslopetest)
assert lnm_check.LeastNonMonotonic(
monoAry, 0) == 0, "LeastNonMonotonic ran unsuccessfully.\n"
# Print results
monoAryflag = 0
for i in range(length):
if monoAry[
i] == 3.0: # Print best monotonic progress variable if it exists
if monoAryflag != 0:
raise RuntimeError(
"Error in contents of monoAry vector: multiple best selected.\n"
)
monoAryflag = 2
bestC[:] = iof.get_progvar(combosmatrix[1:, i], testspecies, locs,
i)
print 'The chosen progress variable is C = %s' % bestC[1][0],
for j in bestC[1][1:]:
print "+ %s" % j,
print '\nThe column numbers of these species are ', bestC[
0], ', respectively.\n'
elif monoAry[
i] == 1.0: # Otherwise print least non-monotonic progress variable
if monoAryflag != 0:
raise RuntimeError("Error in contents of monoAry vector.\n")
monoAryflag = 1
bestC[:] = iof.get_progvar(combosmatrix[1:, i], testspecies, locs,
i)
print 'WARNING: no monotonic progress variables found, but proceeding with best alternative.\n'
print 'The least non-monotonic progress variable is C = %s' % bestC[
1][0],
for j in bestC[1][1:]:
print "+ %s" % j,
print '\nThe column numbers of these species are', bestC[
0], ', respectively.\n'
for i in range(
length): # Print/identify other monotonic progress variables
if monoAry[i] == 2.0:
if monoAryflag < 2:
raise RuntimeError("Error in contents of monoAry vector.\n")
elif monoAryflag == 2:
print "Other candidate monotonic progress variables are:"
otherC = iof.get_progvar(combosmatrix[1:, i], testspecies, locs, i)
print 'C = %s' % otherC[1][0],
for j in otherC[1][1:]:
print "+ %s" % j,
print "\n",
monoAryflag = 3
if monoAryflag < 1: # Give error if no best progress variable is found
raise RuntimeError("Error: no best progress variable selected.")
# Write results
for i in range(nofiles):
filesmatC.SetVal(i, 0, progVar.GetVal(i, bestC[2]))
# Plot results
Cst = np.zeros(nofiles)
Tst = np.zeros(nofiles)
plt.figure()
for ii in range(nofiles):
Tst[ii] = progVar.GetVal(ii, 0)
if options["PlotAllC"][0] == 'yes':
for jj in range(length - 1):
for ii in range(nofiles):
Cst[ii] = progVar.GetVal(ii, jj + 1)
Cst = Cst / Cst.max()
otherplt, = plt.plot(Tst, Cst, color='r')
for ii in range(nofiles):
Cst[ii] = filesmatC.GetVal(ii, 0)
Cst = Cst / Cst.max()
bestplt, = plt.plot(Tst,
Cst,
color='k',
marker='o',
markerfacecolor='none')
if options["PlotAllC"][0] == 'yes':
plt.legend(
[bestplt, otherplt],
['Best progress variable', 'Other candidate progress variables'],
loc='lower right')
plt.xlabel("T (K)")
plt.ylabel("Normalized Progress Variable (C/Cmax)")
plt.title("Best Progress Variable")
if skip == 'yes':
plt.title("User-selected progress variable")
plt.savefig("output/%s.pdf" % 'CvsTemp')
plt.clf()
return filesmatC
inputs,
minargs=0,
default=['simple'])
options["gridextrapolate"] = iof.read_input("extrapolate in fittogrid:",
inputs,
minargs=0,
default=['no'])
# find best progress variable
bestC = []
nofiles = len(datafiles) # Check to see if used later and maybe move
filesmatC = fpv.findC(datafiles, testspecies, bestC, options)
# sort FILESMATRIX by progress variable
if options["sort method"][0] == 'bubble':
sorter = sorting.BubbleSort(filesmatC)
elif options["sort method"][0] == 'standard':
sorter = sorting.StandardSort(filesmatC)
elif options["sort method"][0] == 'brute': # has speed issues
sorter = sorting.BruteSort(filesmatC)
else:
raise IOError(
"invalid sorting method (%s) specified, instead use <bubble> sort" %
sortmethod)
sorter.SetRefColNum(0)
sorter.sort_data()
print "\nSorting filesmatrix by C using %s sort" % options["sort method"][0]
# Calculate PDF matrix
# Get user inputs
Z = np.genfromtxt(datafiles[0],