def build(self, data = None, attributeSet = None):
if attributeSet is None:
# build an attribute set.
aa = AttributeSet()
for line in data:
if line[0] == '@':
aa.add(Attribute(line[1:].split('\t')))
return aa
else:
# build an example set.
ee = ExampleSet()
for line in data:
if type(line) == type(str()):
if line[0] == '#':
ee.add(Example(line[1:], attributeSet))
elif type(line) == type(ExampleSet()):
ee.add(line)
return ee
def __init__(self, filePath=None): self.name = None self.attributes = None self.examples = ExampleSet() self.iteration_index = 0 if filePath is not None: self.initialize(filePath)
def build(self, data=None, attributeSet=None):
""" return an AttributeSet or ExampleSet object
@param data: input data; raw (textual) attributes or examples
@param attributeSet: AttributeSet object required to create ExampleSet objects
@return AttributeSet or ExampleSet objects
"""
# Build an AttributeSet object from raw (text) attributes.
if attributeSet is None:
attributeSet = AttributeSet()
for line in data:
# If the line is prefixed with '@', create an Attribute object and add it to the AttributeSet
if line[0] == '@':
attributeSet.add(Attribute(line[1:].split('\t')))
return attributeSet
# Build an ExampleSet object from raw (text) examples and an AttributeSet.
else:
exampleSet = ExampleSet()
# Loop through the data split by newline
for line in data:
# If the line is a string, check it is an example (prefixed by '#')
if type(line) == type(str()):
# If the line is an example, create an Example object and add it to the ExampleSet
if line[0] == '#':
exampleSet.add(Example(line[1:], attributeSet))
# Commented out for the time being 7/13/2016
#else:
# exampleSet.add(line)
return exampleSet
def getExamplesByClass(self, i = None): """ return examples with label i """ return ExampleSet(self.examples.getExamples(i))
