def sampling(self, nx, xlimits, method='LHS', random_seed=10000):
'''
create nx samples bounded by xlimits using specified method.
xlimits defines lb and ub, in np.array([[LB1, UB1], [LB2, UB2], ...]) format.
method = 'LHS': Latin hypercube sampling, 'CCD': centralized composite design,
'PBD': Plackett-Burman design, 'PB-CCD': Plackett-Burman centralized composite design
'''
n_var = xlimits.shape[0]
# Sampling
if method.lower() == 'lhs':
x = DOE.lhs(n_var,
samples=nx,
criterion='correlation',
random_state=random_seed) * 2.0 - 1.0
elif method.lower() == 'ccd':
if n_var > 8:
raise ValueError(
'number of variables is TOO LARGE for centralized composite design (CCD).'
)
if n_var > 7:
warnings.warn(
'number of variables is TOO LARGE for centralized composite design (CCD).'
)
x = DOE.ccdesign(n_var,
center=(0, 1),
alpha='rotatable',
face='inscribed')
elif method.lower() == 'pbd':
x = DOE.pbdesign(n_var)
elif method.lower() in ['pb-ccd', 'pbccd']:
l = np.sqrt(n_var)
x = DOE.pbdesign(n_var) / l
x = np.append(x, -x / 2.0, axis=0)
for idx in range(0, n_var):
z = np.zeros((1, n_var))
z[0, idx] = 1.0
x = np.append(x, z, axis=0)
z[0, idx] = -1.0
x = np.append(x, z, axis=0)
x = np.append(x, np.zeros((1, n_var)), axis=0)
# Scale
for idx in range(0, xlimits.shape[0]):
x[:, idx] = (x[:, idx] + 1.0) / 2.0 * (
xlimits[idx, 1] - xlimits[idx, 0]) + xlimits[idx, 0]
# Return
return x
def genDoe(nDV, doeMethod):
if doeMethod == 'lhs':
doeTable = DOE.lhs(nDV, samples=nDV * 10)
elif doeMethod == 'pbdesign':
doeTable = DOE.pbdesign(nDV)
elif doeMethod == 'bbdesign':
doeTable = DOE.bbdesign(nDV)
else:
print("Error, You have to check a parameter")
return doeTable
def pbd(self):
"""
Get experiments corresponding to a Plackett-Burman design.
"""
pb = pd.DataFrame(pbdesign(len(self.levels)), columns=self.names)
pb = Data(pb)
pb.standardize(scaler='minmax', target=None)
self.design = pb.data
def genDoe(self):
if self.doeMethod == 'lhs':
doeTable = DOE.lhs(self.nDV, samples=self.nDV * 10)
elif self.doeMethod == 'pbdesign':
doeTable = DOE.pbdesign(self.nDV)
temp = len(doeTable) - 1
for j in range(self.nDV):
doeTable[temp, j] = 0
elif self.doeMethod == 'bbdesign':
doeTable = DOE.bbdesign(self.nDV, center=1)
else:
print("Error, You have to check a parameter")
return doeTable
def _generate_design(self, size):
"""
Generate a Plackett-Burman DOE design.
Parameters
----------
size : int
The number of factors for the design.
Returns
-------
ndarray
The design matrix as a size x levels array of indices.
"""
doe = pyDOE2.pbdesign(size)
doe[doe < 0] = 0 # replace -1 with zero
return doe
def opmsens_write_cases(basefile, header, factors, scenario):
""" Main Function for Writing out Scenario Cases
This is the main function that controls the writing out of the various requested scenario cases (jobs). The
function first calls the opmsens_checkerr routine to check for errors and then the opmsens_clean routine to
remove previously created scenario files. After which the opmsens_write_param and opmsens_write_data functions are
called to create the scenario PARAM and DATA files
Parameters
----------
basefile : str
The basefile used to generate all the cases
header : list
A list of of header names
factors : table
A table of design factors
scenario : str
The type of scenario to be generated
Returns
------
None
"""
# Check for Errors and Return if Errors Found
checkerr = opmsens_check(basefile, header, factors, scenario)
if checkerr:
return ()
#
# Cleanup Existing Files
#
opmsens_clean(basefile)
#
# Define Factor and Job Data Frame
#
df = pd.DataFrame(factors, columns=header)
df = df[df != ''].dropna()
jobdf = pd.DataFrame()
jobdf[header[1]] = df[header[1]]
for slevel in ['Low', 'Best', 'High']:
if slevel in scenario:
jobdf[slevel] = df[slevel]
nfactor = jobdf.shape[0]
nlevel = jobdf.shape[1]
#
# Write PARAM and DATA Files
#
jobs = []
jobstart = 1
joberr = False
jobdata = Path(basefile)
jobparam = Path(basefile).with_suffix('.param')
jobque = Path(basefile).with_suffix('.que')
print('Scenario: ' + scenario + ' Start')
#
# Low, Best and High Scenario
#
if 'Scenario' in scenario:
jobnum = 0
for joblevel in range(1, nlevel):
(joberr, jobs) = opmsens_write_param(jobstart, jobnum, jobparam,
jobdata, jobs)
if joberr:
break
joberr = opmsens_write_data(scenario, joblevel, nfactor, jobdf,
jobstart, jobnum, jobdata)
if joberr:
break
jobstart = jobstart + joblevel
#
# One Job per Factor
#
elif 'One Job per Factor' in scenario:
for joblevel in range(1, nlevel):
for jobnum in range(0, nfactor):
(joberr, jobs) = opmsens_write_param(jobstart, jobnum,
jobparam, jobdata, jobs)
if joberr:
break
joberr = opmsens_write_data(scenario, joblevel, nfactor, jobdf,
jobstart, jobnum, jobdata)
if joberr:
break
jobstart = jobstart + nfactor
#
# Factorial Low and High Full
#
elif 'Factorial' in scenario:
#
# Obtain DOE Matrix and Convert to Data Frame
#
doedata = pd.DataFrame()
if 'Factorial Low and High Full' in scenario:
doedata = pyDOE2.ff2n(nfactor) + 2
if 'Factorial Low and High Plackett-Burman' in scenario:
doedata = pyDOE2.pbdesign(nfactor) + 2
if 'Factorial Low, Best and High Full' in scenario:
doedata = (pyDOE2.fullfact([nlevel - 1] * nfactor)) - 1
if 'Factorial Low, Best and High Box-Behnken' in scenario:
doedata = pyDOE2.bbdesign(nfactor)
doedf = pd.DataFrame(data=doedata).transpose()
doedf = doedf.rename(columns=lambda x: 'RUN' + str(x + 1).zfill(3),
inplace=False)
#
# Set Factor Values
#
for n in range(0, nfactor):
doedf.iloc[n, :] = doedf.iloc[n, :].replace(
[1.0, 2.0, 3.0], [df.iloc[n, 2], df.iloc[n, 3], df.iloc[n, 4]])
#
# Merge Data Frames and Write Out Files
#
jobdf = pd.DataFrame()
jobdf[header[1]] = df[header[1]]
jobdf = pd.concat([jobdf, doedf], axis=1)
nfactor = jobdf.shape[0]
nlevel = jobdf.shape[1]
jobstart = 0
for joblevel in range(1, nlevel):
jobnum = joblevel
(joberr, jobs) = opmsens_write_param(jobstart, jobnum, jobparam,
jobdata, jobs)
if joberr:
break
joberr = opmsens_write_data(scenario, joblevel, nfactor, jobdf,
jobstart, jobnum, jobdata)
if joberr:
break
print('Scenario: ' + scenario + ' End')
if not joberr:
print('WriteQueu: Start')
opmsens_write_queue(jobs)
print('WriteQueu: End')
return ()