Can i possible create cross validation to ID3 algorithm in accord.net?

Question

A snapshot of my code: (full version: http://pastebin.com/7ALhSKgX)

        var crossvalidation = new CrossValidation(size: data.Rows.Count, folds: 7);

        crossvalidation.Fitting = 
             delegate(int k, int[] indicesTrain, int[] indicesValidation)
        {
            //omitted declarations for clarity
            DecisionTree tree = new DecisionTree(attributes, classCount);

            //omitted
            double trainingError = 
               id3learning.ComputeError(trainingInputs, trainingOutputs);
            double validationError = 
               id3learning.ComputeError(validationInputs, validationOutputs);
            return new CrossValidationValues<DecisionTree>
               (tree, trainingError, validationError);
        };

The error is on this line:

          return new CrossValidationValues<DecisionTree>
                        (tree, trainingError, validationError);

and its giving an error : Cannot convert anonymous method to delegate type 'Accord.MachineLearning.CrossValidationFittingFunction' because some of the return types in the block are not implicitly convertible to the delegate return type

It seems your return type is not of the same type that's compatible by watherver .Fitting is expecting. — Stefan, May 07 '14 at 23:40

score 1 · Answer 1 · answered May 08 '14 at 06:02

The problem is that you are using the non-generic constructor CrossValidation to initialize the crossvalidation variable. The CrossValidation class inherits from CrossValidation<object>.

The Fitting property is a CrossValidationFittingFunction<TModel> delegate, where TModel for the non-generic CrossValidation class is object rather than DecisionTree.

Depending on your intentions, you could solve this either by using a more specific constructor:

var crossvalidation = new CrossValidation<DecisionTree>(size: data.Rows.Count, folds: 7);

or return less specific cross-validation values:

return new CrossValidationValues<object>(tree, trainingError, validationError);

score 0 · Answer 2 · answered Aug 22 '17 at 19:24

As of version 3.7.0, it is now possible to create use Cross-Validation without having to write your own Fitting function. An example is shown below:

// Ensure we have reproducible results
Accord.Math.Random.Generator.Seed = 0;

// Get some data to be learned: Here we will download and use Wiconsin's
// (Diagnostic) Breast Cancer dataset, where the goal is to determine
// whether the characteristics extracted from a breast cancer exam
// correspond to a malignant or benign type of cancer. In order to do
// this using the Accord.NET Framework, all we have to do is:
var data = new WisconsinDiagnosticBreastCancer();

// Now, we can import the input features and output labels using
double[][] input = data.Features; // 569 samples, 30-dimensional features
int[] output = data.ClassLabels;  // 569 samples, 2 different class labels

// Now, let's say we want to measure the cross-validation performance of
// a decision tree with a maximum tree height of 5 and where variables
// are able to join the decision path at most 2 times during evaluation:
var cv = CrossValidation.Create(

    k: 10, // We will be using 10-fold cross validation

    learner: (p) => new C45Learning() // here we create the learning algorithm
    {
        Join = 2,
        MaxHeight = 5
    },

    // Now we have to specify how the tree performance should be measured:
    loss: (actual, expected, p) => new ZeroOneLoss(expected).Loss(actual),

    // This function can be used to perform any special
    // operations before the actual learning is done, but
    // here we will just leave it as simple as it can be:
    fit: (teacher, x, y, w) => teacher.Learn(x, y, w),

    // Finally, we have to pass the input and output data
    // that will be used in cross-validation. 
    x: input, y: output
);

// After the cross-validation object has been created,
// we can call its .Learn method with the input and 
// output data that will be partitioned into the folds:
var result = cv.Learn(input, output);

// We can grab some information about the problem:
int numberOfSamples = result.NumberOfSamples; // should be 569
int numberOfInputs = result.NumberOfInputs;   // should be 30
int numberOfOutputs = result.NumberOfOutputs; // should be 2

double trainingError = result.Training.Mean; // should be 0
double validationError = result.Validation.Mean; // should be 0.089661654135338359

Can i possible create cross validation to ID3 algorithm in accord.net?

2 Answers2