Package ML :: Package DecTree :: Module CrossValidate

Module CrossValidate

handles doing cross validation with decision trees

This is, perhaps, a little misleading.  For the purposes of this module,
cross validation == evaluating the accuracy of a tree.

Functions

[hide private]

ChooseOptimalRoot(examples, trainExamples, testExamples, attrs, nPossibleVals, treeBuilder, nQuantBounds=[], **kwargs)
loops through all possible tree roots and chooses the one which produces the best tree **Arguments** - examples: the full set of examples - trainExamples: the training examples - testExamples: the testing examples - attrs: a list of attributes to consider in the tree building - nPossibleVals: a list of the number of possible values each variable can adopt - treeBuilder: the function to be used to actually build the tree - nQuantBounds: an optional list. source code

CrossValidate(tree, testExamples, appendExamples=0)
Determines the classification error for the testExamples **Arguments** - tree: a decision tree (or anything supporting a _ClassifyExample()_ method) - testExamples: a list of examples to be used for testing - appendExamples: a toggle which is passed along to the tree as it does the classification.

source code

CrossValidationDriver(examples, attrs, nPossibleVals, holdOutFrac=0.3, silent=0, calcTotalError=0, treeBuilder=<function ID3Boot at 0x9f7ad4c>, lessGreedy=0, startAt=None, nQuantBounds=[], maxDepth=-1, **kwargs)
Driver function for building trees and doing cross validation **Arguments** - examples: the full set of examples - attrs: a list of attributes to consider in the tree building - nPossibleVals: a list of the number of possible values each variable can adopt - holdOutFrac: the fraction of the data which should be reserved for the hold-out set (used to calculate the error) - silent: a toggle used to control how much visual noise this makes as it goes. source code

TestRun()
testing code...

source code

Variables

[hide private]

Complex0 = 'F'

Complex16 = 'F'

Complex32 = 'F'

Complex64 = 'D'

Complex8 = 'F'

Float0 = 'f'

Float16 = 'f'

Float32 = 'f'

Float64 = 'd'

Float8 = 'f'

Int0 = '1'

Int16 = 's'

Int32 = 'i'

Int8 = '1'

absolute = <ufunc 'absolute'>

add = <ufunc 'add'>

arccos = <ufunc 'arccos'>

arccosh = <ufunc 'arccosh'>

arcsin = <ufunc 'arcsin'>

arcsinh = <ufunc 'arcsinh'>

arctan = <ufunc 'arctan'>

arctan2 = <ufunc 'arctan2'>

arctanh = <ufunc 'arctanh'>

bitwise_and = <ufunc 'bitwise_and'>

bitwise_or = <ufunc 'bitwise_or'>

bitwise_xor = <ufunc 'bitwise_xor'>

ceil = <ufunc 'ceil'>

conjugate = <ufunc 'conjugate'>

cos = <ufunc 'cos'>

cosh = <ufunc 'cosh'>

divide = <ufunc 'divide'>

divide_safe = <ufunc 'divide_safe'>

e = 2.71828182846

equal = <ufunc 'equal'>

exp = <ufunc 'exp'>

fabs = <ufunc 'fabs'>

floor = <ufunc 'floor'>

floor_divide = <ufunc 'floor_divide'>

fmod = <ufunc 'fmod'>

greater = <ufunc 'greater'>

greater_equal = <ufunc 'greater_equal'>

hypot = <ufunc 'hypot'>

invert = <ufunc 'invert'>

left_shift = <ufunc 'left_shift'>

less = <ufunc 'less'>

less_equal = <ufunc 'less_equal'>

log = <ufunc 'log'>

log10 = <ufunc 'log10'>

logical_and = <ufunc 'logical_and'>

logical_not = <ufunc 'logical_not'>

logical_or = <ufunc 'logical_or'>

logical_xor = <ufunc 'logical_xor'>

maximum = <ufunc 'maximum'>

minimum = <ufunc 'minimum'>

multiply = <ufunc 'multiply'>

negative = <ufunc 'negative'>

not_equal = <ufunc 'not_equal'>

pi = 3.14159265359

power = <ufunc 'power'>

remainder = <ufunc 'remainder'>

right_shift = <ufunc 'right_shift'>

sin = <ufunc 'sin'>

sinh = <ufunc 'sinh'>

sqrt = <ufunc 'sqrt'>

subtract = <ufunc 'subtract'>

tan = <ufunc 'tan'>

tanh = <ufunc 'tanh'>

true_divide = <ufunc 'true_divide'>

Function Details

[hide private]

ChooseOptimalRoot(examples, trainExamples, testExamples, attrs, nPossibleVals, treeBuilder, nQuantBounds=`[]`, **kwargs)

source code

loops through all possible tree roots and chooses the one which produces the best tree

**Arguments**

  - examples: the full set of examples

  - trainExamples: the training examples

  - testExamples: the testing examples

  - attrs: a list of attributes to consider in the tree building

  - nPossibleVals: a list of the number of possible values each variable can adopt

  - treeBuilder: the function to be used to actually build the tree

  - nQuantBounds: an optional list.  If present, it's assumed that the builder
    algorithm takes this argument as well (for building QuantTrees)

**Returns**

  The best tree found
  
**Notes**

  1) Trees are built using _trainExamples_

  2) Testing of each tree (to determine which is best) is done using _CrossValidate_ and
     the entire set of data (i.e. all of _examples_)

  3) _trainExamples_ is not used at all, which immediately raises the question of
     why it's even being passed in

CrossValidate(tree, testExamples, appendExamples=0)

source code

Determines the classification error for the testExamples

**Arguments**

  - tree: a decision tree (or anything supporting a _ClassifyExample()_ method)

  - testExamples: a list of examples to be used for testing

  - appendExamples: a toggle which is passed along to the tree as it does
    the classification. The trees can use this to store the examples they
    classify locally.

**Returns**

  a 2-tuple consisting of:

    1) the percent error of the tree

    2) a list of misclassified examples

CrossValidationDriver(examples, attrs, nPossibleVals, holdOutFrac=0.3, silent=0, calcTotalError=0, treeBuilder=<function ID3Boot at 0x9f7ad4c>, lessGreedy=0, startAt=None, nQuantBounds=`[]`, maxDepth=-1, **kwargs)

source code

Driver function for building trees and doing cross validation

**Arguments**

  - examples: the full set of examples

  - attrs: a list of attributes to consider in the tree building

  - nPossibleVals: a list of the number of possible values each variable can adopt

  - holdOutFrac: the fraction of the data which should be reserved for the hold-out set
     (used to calculate the error)

  - silent: a toggle used to control how much visual noise this makes as it goes.

  - calcTotalError: a toggle used to indicate whether the classification error
    of the tree should be calculated using the entire data set (when true) or just
    the training hold out set (when false)

  - treeBuilder: the function to call to build the tree

  - lessGreedy: toggles use of the less greedy tree growth algorithm (see
    _ChooseOptimalRoot_).

  - startAt: forces the tree to be rooted at this descriptor

  - nQuantBounds: an optional list.  If present, it's assumed that the builder
    algorithm takes this argument as well (for building QuantTrees)

  - maxDepth: an optional integer.  If present, it's assumed that the builder
    algorithm takes this argument as well

**Returns**

   a 2-tuple containing:

     1) the tree

     2) the cross-validation error of the tree

TestRun()

source code

testing code

Module CrossValidate

ChooseOptimalRoot(examples, trainExamples, testExamples, attrs, nPossibleVals, treeBuilder, nQuantBounds=[], **kwargs)

CrossValidate(tree, testExamples, appendExamples=0)

CrossValidationDriver(examples, attrs, nPossibleVals, holdOutFrac=0.3, silent=0, calcTotalError=0, treeBuilder=<function ID3Boot at 0x9f7ad4c>, lessGreedy=0, startAt=None, nQuantBounds=[], maxDepth=-1, **kwargs)

TestRun()

ChooseOptimalRoot(examples, trainExamples, testExamples, attrs, nPossibleVals, treeBuilder, nQuantBounds=`[]`, **kwargs)

CrossValidationDriver(examples, attrs, nPossibleVals, holdOutFrac=0.3, silent=0, calcTotalError=0, treeBuilder=<function ID3Boot at 0x9f7ad4c>, lessGreedy=0, startAt=None, nQuantBounds=`[]`, maxDepth=-1, **kwargs)