Metrics for Machine Learning evaluation Data Science Measurement

## Project description

Metrics for evaluating machine learning models or Data Science

Include : All metrics from SKLEARN. Category based metrics. ########################################################################

from metric.metric import *

##### Classification metrics

accuracy_score(y_true,Â y_pred[,Â â€¦]) Accuracy classification score. auc(x,Â y) Compute Area Under the Curve (AUC) using the trapezoidal rule average_precision_score(y_true,Â y_score) Compute average precision (AP) from prediction scores balanced_accuracy_score(y_true,Â y_pred) Compute the balanced accuracy brier_score_loss(y_true,Â y_prob[,Â â€¦]) Compute the Brier score. classification_report(y_true,Â y_pred) Build a text report showing the main classification metrics cohen_kappa_score(y1,Â y2[,Â labels,Â â€¦]) Cohenâ€™s kappa: a statistic that measures inter-annotator agreement. confusion_matrix(y_true,Â y_pred[,Â â€¦]) Compute confusion matrix to evaluate the accuracy of a classification. dcg_score(y_true,Â y_score[,Â k,Â â€¦]) Compute Discounted Cumulative Gain. f1_score(y_true,Â y_pred[,Â labels,Â â€¦]) Compute the F1 score, also known as balanced F-score or F-measure fbeta_score(y_true,Â y_pred,Â beta[,Â â€¦]) Compute the F-beta score hamming_loss(y_true,Â y_pred[,Â â€¦]) Compute the average Hamming loss. hinge_loss(y_true,Â pred_decision[,Â â€¦]) Average hinge loss (non-regularized) jaccard_score(y_true,Â y_pred[,Â â€¦]) Jaccard similarity coefficient score log_loss(y_true,Â y_pred[,Â eps,Â â€¦]) Log loss, aka logistic loss or cross-entropy loss. matthews_corrcoef(y_true,Â y_pred[,Â â€¦]) Compute the Matthews correlation coefficient (MCC) multilabel_confusion_matrix(y_true,Â â€¦) Compute a confusion matrix for each class or sample ndcg_score(y_true,Â y_score[,Â k,Â â€¦]) Compute Normalized Discounted Cumulative Gain. precision_recall_curve(y_true,Â â€¦) Compute precision-recall pairs for different probability thresholds precision_recall_fscore_support(â€¦) Compute precision, recall, F-measure and support for each class precision_score(y_true,Â y_pred[,Â â€¦]) Compute the precision recall_score(y_true,Â y_pred[,Â â€¦]) Compute the recall roc_auc_score(y_true,Â y_score[,Â â€¦]) Compute Area Under the Receiver Operating Characteristic Curve (ROC AUC) from prediction scores. roc_curve(y_true,Â y_score[,Â â€¦]) Compute Receiver operating characteristic (ROC) zero_one_loss(y_true,Â y_pred[,Â â€¦]) Zero-one classification loss.

##### Regression metrics

explained_variance_score(y_true,Â y_pred) Explained variance regression score function max_error(y_true,Â y_pred) max_error metric calculates the maximum residual error. mean_absolute_error(y_true,Â y_pred) Mean absolute error regression loss mean_squared_error(y_true,Â y_pred[,Â â€¦]) Mean squared error regression loss mean_squared_log_error(y_true,Â y_pred) Mean squared logarithmic error regression loss median_absolute_error(y_true,Â y_pred) Median absolute error regression loss r2_score(y_true,Â y_pred[,Â â€¦]) R^2 (coefficient of determination) regression score function. mean_poisson_deviance(y_true,Â y_pred) Mean Poisson deviance regression loss. mean_gamma_deviance(y_true,Â y_pred) Mean Gamma deviance regression loss. mean_tweedie_deviance(y_true,Â y_pred) Mean Tweedie deviance regression loss.

##### Multilabel ranking metrics

coverage_error(y_true,Â y_score[,Â â€¦]) Coverage error measure label_ranking_average_precision_score(â€¦) Compute ranking-based average precision label_ranking_loss(y_true,Â y_score) Compute Ranking loss measure

##### Clustering metrics

supervised, which uses a ground truth class values for each sample. unsupervised, which does not and measures the â€˜qualityâ€™ of the model itself.

adjusted_mutual_info_score(â€¦[,Â â€¦]) Adjusted Mutual Information between two clusterings. adjusted_rand_score(labels_true,Â â€¦) Rand index adjusted for chance. calinski_harabasz_score(X,Â labels) Compute the Calinski and Harabasz score. davies_bouldin_score(X,Â labels) Computes the Davies-Bouldin score. completeness_score(labels_true,Â â€¦) Completeness metric of a cluster labeling given a ground truth. cluster.contingency_matrix(â€¦[,Â â€¦]) Build a contingency matrix describing the relationship between labels. fowlkes_mallows_score(labels_true,Â â€¦) Measure the similarity of two clusterings of a set of points. homogeneity_completeness_v_measure(â€¦) Compute the homogeneity and completeness and V-Measure scores at once. homogeneity_score(labels_true,Â â€¦) Homogeneity metric of a cluster labeling given a ground truth. mutual_info_score(labels_true,Â â€¦) Mutual Information between two clusterings. normalized_mutual_info_score(â€¦[,Â â€¦]) Normalized Mutual Information between two clusterings. silhouette_score(X,Â labels[,Â â€¦]) Compute the mean Silhouette Coefficient of all samples. silhouette_samples(X,Â labels[,Â metric]) Compute the Silhouette Coefficient for each sample. v_measure_score(labels_true,Â labels_pred) V-measure cluster labeling given a ground truth.

Biclustering metrics

consensus_score(a,Â b[,Â similarity]) The similarity of two sets of biclusters.

Pairwise metrics

pairwise.additive_chi2_kernel(X[,Â Y]) Computes the additive chi-squared kernel between observations in X and Y pairwise.chi2_kernel(X[,Â Y,Â gamma]) Computes the exponential chi-squared kernel X and Y. pairwise.cosine_similarity(X[,Â Y,Â â€¦]) Compute cosine similarity between samples in X and Y. pairwise.cosine_distances(X[,Â Y]) Compute cosine distance between samples in X and Y. pairwise.distance_metrics() Valid metrics for pairwise_distances. pairwise.euclidean_distances(X[,Â Y,Â â€¦]) Considering the rows of X (and Y=X) as vectors, compute the distance matrix between each pair of vectors. pairwise.haversine_distances(X[,Â Y]) Compute the Haversine distance between samples in X and Y pairwise.kernel_metrics() Valid metrics for pairwise_kernels pairwise.laplacian_kernel(X[,Â Y,Â gamma]) Compute the laplacian kernel between X and Y. pairwise.linear_kernel(X[,Â Y,Â â€¦]) Compute the linear kernel between X and Y. pairwise.manhattan_distances(X[,Â Y,Â â€¦]) Compute the L1 distances between the vectors in X and Y. pairwise.nan_euclidean_distances(X) Calculate the euclidean distances in the presence of missing values. pairwise.pairwise_kernels(X[,Â Y,Â â€¦]) Compute the kernel between arrays X and optional array Y. pairwise.polynomial_kernel(X[,Â Y,Â â€¦]) Compute the polynomial kernel between X and Y. pairwise.rbf_kernel(X[,Â Y,Â gamma]) Compute the rbf (gaussian) kernel between X and Y. pairwise.sigmoid_kernel(X[,Â Y,Â â€¦]) Compute the sigmoid kernel between X and Y. pairwise.paired_euclidean_distances(X,Â Y) Computes the paired euclidean distances between X and Y pairwise.paired_manhattan_distances(X,Â Y) Compute the L1 distances between the vectors in X and Y. pairwise.paired_cosine_distances(X,Â Y) Computes the paired cosine distances between X and Y pairwise.paired_distances(X,Â Y[,Â metric]) Computes the paired distances between X and Y. pairwise_distances(X[,Â Y,Â metric,Â â€¦]) Compute the distance matrix from a vector array X and optional Y. pairwise_distances_argmin(X,Â Y[,Â â€¦]) Compute minimum distances between one point and a set of points. pairwise_distances_argmin_min(X,Â Y) Compute minimum distances between one point and a set of points. pairwise_distances_chunked(X[,Â Y,Â â€¦]) Generate a distance matrix chunk by chunk with optional reduction