expectedShape = self.Inputs[0].meta.shape
for slot in self.Inputs:
assert slot.meta.shape == expectedShape, "Can't add images of different shapes!"
self.Output.meta.assignFrom(self.Inputs[0].meta)
def execute(self, slot, subindex, roi, result):
result[...] = numpy.zeros(result.shape)
for slot in self.Inputs:
result[...] += slot.get(roi).wait()
return result
from lazyflow.operators import OpArrayPiper
from lazyflow.graph import OperatorWrapper
opGenericMultiOut = OperatorWrapper(OpArrayPiper, graph=graph)
opGenericMultiOut.Input.resize(3)
opGenericMultiOut.name = "OpSomeProvider"
svgOp1 = SvgOperator(opGenericMultiOut, max_child_depth=0)
opSum = OpMultiSum(graph=graph)
opSum.Inputs.resize(3)
svgOp = SvgOperator(opSum, max_child_depth=1)
class OpThreshold(Operator):
ThresholdLevel = InputSlot()
Input = InputSlot()
Output = OutputSlot()
def setupOutputs(self):
self.Output.meta.assignFrom(self.Input.meta)
self.Output.meta.dtype = numpy.uint8
def __init__(self):
# Create a graph to be shared by all operators
graph = Graph()
super(PixelClassificationWorkflow, self).__init__( graph=graph )
self._applets = []
######################
# Interactive workflow
######################
## Create applets
self.projectMetadataApplet = ProjectMetadataApplet()
self.dataSelectionApplet = DataSelectionApplet(self, "Input Data", "Input Data", supportIlastik05Import=True, batchDataGui=False)
self.featureSelectionApplet = FeatureSelectionApplet(self, "Feature Selection", "FeatureSelections")
self.pcApplet = PixelClassificationApplet(self, "PixelClassification")
## Access applet operators
opData = self.dataSelectionApplet.topLevelOperator
opTrainingFeatures = self.featureSelectionApplet.topLevelOperator
opClassify = self.pcApplet.topLevelOperator
## Connect operators ##
# Input Image -> Feature Op
# and -> Classification Op (for display)
opTrainingFeatures.InputImage.connect( opData.Image )
opClassify.InputImages.connect( opData.Image )
# Feature Images -> Classification Op (for training, prediction)
opClassify.FeatureImages.connect( opTrainingFeatures.OutputImage )
opClassify.CachedFeatureImages.connect( opTrainingFeatures.CachedOutputImage )
# Training flags -> Classification Op (for GUI restrictions)
opClassify.LabelsAllowedFlags.connect( opData.AllowLabels )
######################
# Batch workflow
######################
## Create applets
self.batchInputApplet = DataSelectionApplet(self, "Batch Inputs", "BatchDataSelection", supportIlastik05Import=False, batchDataGui=True)
self.batchResultsApplet = BatchIoApplet(self, "Batch Results")
## Access applet operators
opBatchInputs = self.batchInputApplet.topLevelOperator
opBatchInputs.name = 'opBatchInputs'
opBatchResults = self.batchResultsApplet.topLevelOperator
## Create additional batch workflow operators
opBatchFeatures = OperatorWrapper( OpFeatureSelection, graph=graph, parent=self, promotedSlotNames=['InputImage'] )
opBatchFeatures.name = "opBatchFeatures"
opBatchPredictor = OperatorWrapper( OpPredictRandomForest, graph=graph, parent=self, promotedSlotNames=['Image'])
opBatchPredictor.name = "opBatchPredictor"
opSelectBatchDatasetPath = OperatorWrapper( OpAttributeSelector, graph=graph, parent=self )
## Connect Operators ##
# Provide dataset paths from data selection applet to the batch export applet via an attribute selector
opSelectBatchDatasetPath.InputObject.connect( opBatchInputs.Dataset )
opSelectBatchDatasetPath.AttributeName.setValue( 'filePath' )
opBatchResults.DatasetPath.connect( opSelectBatchDatasetPath.Result )
# Connect (clone) the feature operator inputs from
# the interactive workflow's features operator (which gets them from the GUI)
opBatchFeatures.Scales.connect( opTrainingFeatures.Scales )
opBatchFeatures.FeatureIds.connect( opTrainingFeatures.FeatureIds )
opBatchFeatures.SelectionMatrix.connect( opTrainingFeatures.SelectionMatrix )
# Classifier and LabelsCount are provided by the interactive workflow
opBatchPredictor.Classifier.connect( opClassify.Classifier )
opBatchPredictor.LabelsCount.connect( opClassify.MaxLabelValue )
# Connect Image pathway:
# Input Image -> Features Op -> Prediction Op -> Export
opBatchFeatures.InputImage.connect( opBatchInputs.Image )
opBatchPredictor.Image.connect( opBatchFeatures.OutputImage )
opBatchResults.ImageToExport.connect( opBatchPredictor.PMaps )
self._applets.append(self.projectMetadataApplet)
self._applets.append(self.dataSelectionApplet)
self._applets.append(self.featureSelectionApplet)
self._applets.append(self.pcApplet)
self._applets.append(self.batchInputApplet)
self._applets.append(self.batchResultsApplet)
# The shell needs a slot from which he can read the list of image names to switch between.
# Use an OpAttributeSelector to create a slot containing just the filename from the OpDataSelection's DatasetInfo slot.
opSelectFilename = OperatorWrapper( OpAttributeSelector, graph=graph, parent=self )
opSelectFilename.InputObject.connect( opData.Dataset )
opSelectFilename.AttributeName.setValue( 'filePath' )
self._imageNameListSlot = opSelectFilename.Result
expectedShape = self.Inputs[0].meta.shape
for slot in self.Inputs:
assert slot.meta.shape == expectedShape, "Can't add images of different shapes!"
self.Output.meta.assignFrom(self.Inputs[0].meta)
def execute(self, slot, subindex, roi, result):
result[...] = numpy.zeros(result.shape)
for slot in self.Inputs:
result[...] += slot.get(roi).wait()
return result
from lazyflow.operators import OpArrayPiper
from lazyflow.graph import OperatorWrapper
opGenericMultiOut = OperatorWrapper( OpArrayPiper, graph=graph )
opGenericMultiOut.Input.resize(3)
opGenericMultiOut.name = "OpSomeProvider"
svgOp1 = SvgOperator(opGenericMultiOut, max_child_depth=0)
opSum = OpMultiSum(graph=graph)
opSum.Inputs.resize(3)
svgOp = SvgOperator(opSum, max_child_depth=1)
class OpThreshold(Operator):
ThresholdLevel = InputSlot()
Input = InputSlot()
Output = OutputSlot()
def setupOutputs(self):
self.Output.meta.assignFrom(self.Input.meta)
self.Output.meta.dtype = numpy.uint8
def __init__(self):
super(PixelClassificationWorkflow, self).__init__()
self._applets = []
# Create a graph to be shared by all operators
graph = Graph()
self._graph = graph
######################
# Interactive workflow
######################
## Create applets
self.projectMetadataApplet = ProjectMetadataApplet()
self.dataSelectionApplet = DataSelectionApplet(
graph,
"Input Data",
"Input Data",
supportIlastik05Import=True,
batchDataGui=False)
self.featureSelectionApplet = FeatureSelectionApplet(
graph, "Feature Selection", "FeatureSelections")
self.pcApplet = PixelClassificationApplet(graph, "PixelClassification")
## Access applet operators
opData = self.dataSelectionApplet.topLevelOperator
opTrainingFeatures = self.featureSelectionApplet.topLevelOperator
opClassify = self.pcApplet.topLevelOperator
## Connect operators ##
# Input Image -> Feature Op
# and -> Classification Op (for display)
opTrainingFeatures.InputImage.connect(opData.Image)
opClassify.InputImages.connect(opData.Image)
# Feature Images -> Classification Op (for training, prediction)
opClassify.FeatureImages.connect(opTrainingFeatures.OutputImage)
opClassify.CachedFeatureImages.connect(
opTrainingFeatures.CachedOutputImage)
# Training flags -> Classification Op (for GUI restrictions)
opClassify.LabelsAllowedFlags.connect(opData.AllowLabels)
######################
# Batch workflow
######################
## Create applets
self.batchInputApplet = DataSelectionApplet(
graph,
"Batch Inputs",
"BatchDataSelection",
supportIlastik05Import=False,
batchDataGui=True)
self.batchResultsApplet = BatchIoApplet(graph, "Batch Results")
## Access applet operators
opBatchInputs = self.batchInputApplet.topLevelOperator
opBatchInputs.name = 'opBatchInputs'
opBatchResults = self.batchResultsApplet.topLevelOperator
## Create additional batch workflow operators
opBatchFeatures = OperatorWrapper(OpFeatureSelection,
graph=graph,
promotedSlotNames=['InputImage'])
opBatchFeatures.name = "opBatchFeatures"
opBatchPredictor = OperatorWrapper(OpPredictRandomForest,
graph=graph,
promotedSlotNames=['Image'])
opBatchPredictor.name = "opBatchPredictor"
opSelectBatchDatasetPath = OperatorWrapper(OpAttributeSelector,
graph=graph)
## Connect Operators ##
# Provide dataset paths from data selection applet to the batch export applet via an attribute selector
opSelectBatchDatasetPath.InputObject.connect(opBatchInputs.Dataset)
opSelectBatchDatasetPath.AttributeName.setValue('filePath')
opBatchResults.DatasetPath.connect(opSelectBatchDatasetPath.Result)
# Connect (clone) the feature operator inputs from
# the interactive workflow's features operator (which gets them from the GUI)
opBatchFeatures.Scales.connect(opTrainingFeatures.Scales)
opBatchFeatures.FeatureIds.connect(opTrainingFeatures.FeatureIds)
opBatchFeatures.SelectionMatrix.connect(
opTrainingFeatures.SelectionMatrix)
# Classifier and LabelsCount are provided by the interactive workflow
opBatchPredictor.Classifier.connect(opClassify.Classifier)
opBatchPredictor.LabelsCount.connect(opClassify.MaxLabelValue)
# Connect Image pathway:
# Input Image -> Features Op -> Prediction Op -> Export
opBatchFeatures.InputImage.connect(opBatchInputs.Image)
opBatchPredictor.Image.connect(opBatchFeatures.OutputImage)
opBatchResults.ImageToExport.connect(opBatchPredictor.PMaps)
self._applets.append(self.projectMetadataApplet)
self._applets.append(self.dataSelectionApplet)
self._applets.append(self.featureSelectionApplet)
self._applets.append(self.pcApplet)
self._applets.append(self.batchInputApplet)
self._applets.append(self.batchResultsApplet)
# The shell needs a slot from which he can read the list of image names to switch between.
# Use an OpAttributeSelector to create a slot containing just the filename from the OpDataSelection's DatasetInfo slot.
opSelectFilename = OperatorWrapper(OpAttributeSelector, graph=graph)
opSelectFilename.InputObject.connect(opData.Dataset)
opSelectFilename.AttributeName.setValue('filePath')
self._imageNameListSlot = opSelectFilename.Result
