return m3
from dataset_reader import Dataset, Testset
# Plots high sample count last image.
parser = optparse.OptionParser()
parser.add_option("-s", "--samples", dest="samples", default=2048, type="int")
parser.add_option("-c", "--distribution", dest="distribution", default=None)
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
# TODO: Temporary
copyfile(dataset.testSet().targetMeshPath, "/tmp/mts_mesh_intensity_slot_0.ply")
renderable = dataset.testSet().renderable(1)
pt0 = np.array(dataset.lastAvailableBSDF())
pt1 = np.array(dataset.testSet().targetBSDF())
paramList = dataset.testSet().parameterList()
#print paramList
#print pt0
#print toMap(paramList, pt1)
if dataset.testSet().bsdfAdaptiveSampled:
adaptiveParamList = dataset.bsdfAdaptiveSamplingParameterList
print adaptiveParamList
adaptiveParamMap = toMap(adaptiveParamList, len(adaptiveParamList) * [1.0])
else:
adaptiveParamMap = {}
parser.add_option("-l", "--linear", dest="linear", action="store_true", default=False)
parser.add_option("-c", "--distribution", dest="distribution", default=None, type="string")
parser.add_option("-r", "--samples", dest="samples", default=128, type="int")
parser.add_option("-d", "--direct-samples", dest="directSamples", default=4, type="int")
parser.add_option("-i", "--iteration", dest="iteration", default=None, type="int")
parser.add_option("-s", "--super-iteration", dest="superiteration", default=0, type="int")
parser.add_option("-v", "--verbose", dest="verbose", action="store_true", default=False)
parser.add_option("--direct-only", dest="directOnly", action="store_true", default=False)
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
testset = dataset.testSet()
if options.directOnly:
idirpath = directory + "/renders/direct-targets"
else:
idirpath = directory + "/renders/indirect-targets"
if not os.path.exists(idirpath):
os.mkdir(idirpath)
meshfile = testset.targetMeshPath
paramlist = testset.parameterList()
bsdf = toMap(paramlist, testset.targetBSDF())
for k in range(dataset.testSet().numLights()):
print "Light ", k, "/", dataset.testSet().numLights(), "\r",
parser.add_option("-v",
"--verbose",
dest="verbose",
action="store_true",
default=False)
parser.add_option("--direct-only",
dest="directOnly",
action="store_true",
default=False)
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
testset = dataset.testSet()
if not options.directOnly:
idirpath = directory + "/renders/indirect"
else:
idirpath = directory + "/renders/direct"
if not os.path.exists(idirpath):
os.mkdir(idirpath)
for i in range(dataset.testSet().numIterations() +
dataset.testSet().numBSDFIterations()):
if i >= dataset.testSet().numIterations():
ii = 0
bi = i - dataset.testSet().numIterations()
else:
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
if options.superIndex == -1 and options.index == -1:
targetBSDF = dataset.lastAvailableBSDF()
else:
targetBSDF = dataset.BSDFAt(iteration=options.index,
superiteration=options.superIndex)
print targetBSDF
if options.dryRun:
sys.exit(0)
testset = dataset.testSet()
sphereXML = os.path.dirname(__file__) + "/data/sphere-embeddable.xml"
sphereXMLP = os.path.dirname(__file__) + "/data/sphere-postprocessed.xml"
testset.embedOnto(sphereXML, sphereXMLP)
renderable = testset.renderables[0]
renderable.setFile(sphereXMLP)
renderable.setEmbeddedParameter("envmap", "doge.exr")
if options.distribution is not None:
localThreads = 2
else:
localThreads = 8
sphereImage = renderable.renderReadback(
parser.add_option("--layout-columns",
dest="layoutColumns",
default=5,
type="int")
parser.add_option("--layout-rows", dest="layoutRows", default=5, type="int")
parser.add_option("--adaptive-mode",
dest="adaptiveMode",
default=None,
type="string")
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
testset = dataset.testSet()
gdirpath = directory + "/images/raw-bsdf-gradients"
lr = options.layoutRows
lc = options.layoutColumns
if not os.path.exists(gdirpath):
os.mkdir(gdirpath)
for k in range(dataset.testSet().numLights()):
xmlAddGradientSliceRendering(testset.gradientRenderables[k],
slices=len(testset.parameterList()))
for i in range(dataset.testSet().numBSDFIterations()):
if options.iteration is not None and i != options.iteration:
# from dataset_reader import Dataset (Alternative temporary fix)
#execfile(os.path.dirname(__file__) + "/../tools/dataset_reader.py")
parser = optparse.OptionParser()
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
profiles = [{
"name":
"invAlpha",
"sampleWeights":
invAlpha(dataset.testSet().bsdfDictionary).tolist(),
"type":
"bsdf-adaptive"
}, {
"name":
"uniform",
"sampleWeights": [1.0] * len(dataset.testSet().bsdfDictionary["elements"]),
"type":
"bsdf-adaptive"
}, {
"name": "bsdfWeight",
"sampleWeights": dataset.lastAvailableBSDF(),
"type": "bsdf-adaptive"
}]
uniform = np.stack(
import merl_io
parser = optparse.OptionParser()
parser.add_option("-l", "--linear", dest="linear", action="store_true", default=False)
parser.add_option("-c", "--distribution", dest="distribution", default=None, type="string")
parser.add_option("-r", "--samples", dest="samples", default=128, type="int")
parser.add_option("-i", "--iteration", dest="iteration", default=None, type="int")
parser.add_option("-s", "--super-iteration", dest="superiteration", default=0, type="int")
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
testset = dataset.testSet()
gdirpath = directory + "/meshes/single-bounce-gradients"
idirpath = directory + "/images/single-bounce-currents"
gidirpath = directory + "/renders/single-bounce-gradients"
if not os.path.exists(gdirpath):
os.mkdir(gdirpath)
if not os.path.exists(idirpath):
os.mkdir(idirpath)
if not os.path.exists(gidirpath):
os.mkdir(gidirpath)
print(dataset.testSet().numIterations() + dataset.testSet().numBSDFIterations())
print("")
print("")
type="string")
parser.add_option("-k",
"--resolution",
dest="resolution",
default=256,
type="int")
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
pt = np.array(dataset.lastAvailableBSDF())
if not dataset.testSet().bsdfAdaptiveSampled:
print("BSDF Adaptive Sampling should be enabled for variance testing")
sys.exit(1)
adaptiveParamList = dataset.bsdfAdaptiveSamplingParameterList
paramList = dataset.testSet().parameterList()
testset = dataset.testSet()
###
# Monte Carlo gradient
###
# override
for k in range(dataset.testSet().numLights()):
testset.gradientRenderables[k].setEmbeddedParameter(
"sampleCount", options.samples)
default=False)
(options, args) = parser.parse_args()
directory = args[0]
dataset = Dataset(directory)
if options.sampleFile is None:
print("Using default sample profile filename")
options.sampleFile = directory + "/spatial_sample_profiles.json"
sampleProfiles = json.load(open(options.sampleFile, "r"))
pt = np.array(dataset.lastAvailableBSDF())
if not dataset.testSet().bsdfAdaptiveSampled:
print("BSDF Adaptive Sampling should be enabled for variance testing")
sys.exit(1)
adaptiveParamList = dataset.bsdfAdaptiveSamplingParameterList
paramList = dataset.testSet().parameterList()
testset = dataset.testSet()
# override
for k in range(dataset.testSet().numLights()):
testset.gradientRenderables[k].setEmbeddedParameter(
"sampleCount", options.renderSamples)
testset.gradientRenderables[k].setParameter("blockSize", 8)
reductors = None