improved updating of reference results
This commit is contained in:
parent
a329a9b91d
commit
7960600dbe
|
@ -17,11 +17,11 @@ class Test():
|
|||
self.dirBase = os.path.dirname(os.path.realpath(sys.argv[0]))
|
||||
|
||||
def execute(self,variants = [],update = []):
|
||||
|
||||
'''
|
||||
Run all variants and report first failure.
|
||||
'''
|
||||
if len(variants) == 0: variants = xrange(len(self.variants)) # iterate over all variants
|
||||
|
||||
self.clean()
|
||||
for variant in variants:
|
||||
try:
|
||||
|
@ -35,11 +35,11 @@ class Test():
|
|||
except:
|
||||
return variant
|
||||
return -1
|
||||
# for variant in variants:
|
||||
# for self.current_variant_index,variant in enumerate(variants):
|
||||
# self.prepare(variant)
|
||||
# self.run(variant)
|
||||
# self.postprocess(variant)
|
||||
# if variant in update:
|
||||
# if self.current_variant_index in update:
|
||||
# self.update(variant)
|
||||
# elif not self.compare(variant):
|
||||
# return variant
|
||||
|
@ -144,10 +144,13 @@ class Test():
|
|||
table = damask.ASCIItable(refFile)
|
||||
table.head_read()
|
||||
refFile.close()
|
||||
refArray = numpy.loadtxt(refName,skiprows = table.headerLen)
|
||||
curArray = numpy.loadtxt(curName,skiprows = table.headerLen)
|
||||
|
||||
refArray = numpy.genfromtxt(refName,missing_values='n/a',skip_header = table.headerLen)
|
||||
curArray = numpy.genfromtxt(curName,missing_values='n/a',skip_header = table.headerLen)
|
||||
err = abs((refArray/curArray)-1.) # relative tolerance
|
||||
refNaN=len(numpy.isnan(refArray))
|
||||
curNaN=len(numpy.isnan(curArray))
|
||||
if curNaN == refNaN:
|
||||
err[numpy.isnan(err)]=0.0
|
||||
max_err = numpy.max(err)
|
||||
print 'maximum relative error',max_err
|
||||
return max_err
|
||||
|
|
Loading…
Reference in New Issue