Compute a symmetric two-dimensional array from a non-symmetric two-dimensional array M by taking the minimum between M and M^T. @param M

Generate bagged ensemble

Return the indexes of all the beginnings of a column for a flatten square matrix of size n. @param n Size of the matrix @return An array o

Compute the weighted matrix from the input array of Vectors.

Compute the distance between two vectors. @param vector @param vector2 @return The distance between the two vectors.

Compute the matrix of the distances between all the vectors of the <i>input</i> array. @return A 2-dimensional array of double of the distances.

Compute a non-symmetric matrix by keeping for each row the distance of the <i>k</i> nearest points, and setting the rest to 0. @param distances @retu

Build a conditional FP-tree from the given prefix paths.

Function which reads from the file and yields a generator

Cross-Validation-routine from svm-train */

Train a given classifier n times and prints his confusion matrix and the accuracy of the classifier with a margin of error (by Chebychev

Find frequent itemsets in the given transactions using FP-growth. This function returns a generator instead of an eagerly-populated list of i

Perform internal cross-validation fit

Compute the Gaussian similarity function for non-zero values. @param d Input value @return Calculated value

get probability

@return the weighted Matrix

Increment the count associated with this node's item.

Join a set with itself and returns the n-element itemsets

Loads the data from file and build the dataset in scikit format. () -> Dataset

Method 2 from the multiclass_prob paper by Wu, Lin, and Weng

create sparse vector of features

from each line remove first (label) and parse rest (features) into a list of (featureID, value) tuples.

prints the generated itemsets sorted by support and the confidence rules sorted by confidence

Return the indexes of all the beginnings of a row for a flatten square matrix of size n. @param n Size of the matrix @return An array of i

Starts the classification Pipeline

run the apriori algorithm. data_iter is a record iterator Return both: - items (tuple, support) - rules ((pretuple, posttuple), con

Platt's binary SVM Probablistic Output: an improvement from Lin et al.

construct and solve various formulations

Function to sort a two-dimensionnal array @param toSort the array to sort @return The sorted array

Given the dataset, split in two datasets: One is the Training set. Other is the Test set. The proportion is 80% to 20% Respectively

Returns non empty subsets of arr

Cross-validation decision values for probability estimates

Stratified cross validation

Return parameter of a Laplace distribution