Machine Learning

• 

  Classification

  Predict a Categorical Response
• 

  Regression

  Predict a Numerical Response
• 

  Time Series

  Predict Dynamically
• 

  Clustering & Anomaly Detection

  Find Clusters and Anomalies
• 

  Decomposition & Preprocessing

  Decompose and Preprocess
• 

  Pipeline

  Create ML Pipelines
• 

  memModel

  Independent ML models
• 

  Examples

  Models Usage Examples

Tools

• 

  Model Selection

  Cross Validate, Train/Test Split...
• 

  Metrics

  F1-Score, AUC, MSE...
• 

  Tools

  Models Import/Export...

API Reference

verticapy.learn.cluster

Class	Definition
BisectingKMeans	Creates a BisectingKMeans object using the Vertica BisectingKMeans algorithm.
DBSCAN	Creates a DBSCAN object using the DBSCAN algorithm as defined by Martin Ester, Hans-Peter Kriegel, Jörg Sander and Xiaowei Xu.
KMeans	Creates a KMeans object using the Vertica k-means algorithm.

verticapy.learn.decomposition

Class	Definition
PCA	Creates a PCA (Principal Component Analysis) object using the Vertica PCA algorithm.
SVD	Creates a SVD (Singular Value Decomposition) object using the Vertica SVD algorithm.

verticapy.learn.ensemble

Class	Definition
IsolationForest	Creates an IsolationForest object using the Vertica IFOREST algorithm.
RandomForestClassifier	Creates a RandomForestClassifier object using the Vertica random forest algorithm.
RandomForestRegressor	Creates a RandomForestRegressor object using the Vertica random forest algorithm.
XGBoostClassifier	Creates a XGBoostClassifier object using the Vertica XGB_CLASSIFIER algorithm.
XGBoostRegressor	Creates a XGBoostRegressor object using the Vertica XGB_REGRESSOR algorithm.

verticapy.learn.linear_model

Class	Definition
ElasticNet	Creates a ElasticNet object using the Vertica linear regression algorithm.
Lasso	Creates a Lasso object using the Verticalinear regression algorithm.
LinearRegression	Creates a LinearRegression object using the Vertica linear regression algorithm.
LogisticRegression	Creates a LogisticRegression object using the Vertica logistic regression algorithm.
Ridge	Creates a Ridge object using the Vertica linear regression algorithm.

verticapy.learn.memmodel

Class	Definition
memModel	Creates platform-independent machine learning models that you can export as SQL or Python code for deployment in other environments.

verticapy.learn.metrics

Function	Definition
accuracy_score	Computes the Accuracy Score.
anova_table	Computes the Anova Table.
auc	Computes the ROC AUC (Area Under Curve).
classification_report / report	Computes a classification report using multiple metrics (AUC, accuracy, PRC AUC, F1...).
confusion_matrix	Computes the Confusion Matrix.
critical_success_index	Computes the Critical Success Index.
explained_variance	Computes the Explained Variance.
f1_score	Computes the F1 Score.
informedness	Computes the Informedness.
log_loss	Computes the Log Loss.
markedness	Computes the Markedness.
matthews_corrcoef	Computes the Matthews Correlation Coefficient.
max_error	Computes the Max Error.
mean_absolute_error	Computes the Mean Absolute Error.
mean_squared_error	Computes the Mean Squared Error.
mean_squared_log_error	Computes the Mean Squared Log Error.
median_absolute_error	Computes the Median Absolute Error.
multilabel_confusion_matrix	Computes the Multi Label Confusion Matrix.
negative_predictive_score	Computes the Negative Predictive Score.
prc_auc	Computes the PRC AUC (Area Under Curve).
precision_score	Computes the Precision Score.
recall_score	Computes the Recall Score.
r2_score	Computes the R2 Score.
regression_report	Computes a regression report using multiple metrics (r2, mse, max error...).
specificity_score	Computes the Specificity Score.

verticapy.learn.model_selection

Function	Definition
autoML	Tests multiple models to find the ones which maximize the input score.
bayesian_search_cv	Computes the k-fold bayesian search of an estimator using a random forest model to estimate a probable optimal set of parameters.
best_k	Finds the k-means k based on a score.
cross_validate	Computes the k-fold cross-validation of an estimator.
elbow	Draws the an elbow curve.
enet_search_cv	Computes the k-fold grid search using multiple enet model.
gen_params_grid	Generates the estimator grid.
grid_search_cv	Computes the k-fold grid search of an estimator.
learning_curve	Draws the learning curve.
lift_chart	Draws a lift chart.
parameter_grid	Generates the list of the different input parameters grid combinations.
plot_acf_pacf	Draws ACF and PACF Charts.
prc_curve	Draws a precision-recall curve.
randomized_features_search_cv	Computes the k-fold grid search of an estimator using different features combinations.
randomized_search_cv	Computes the k-fold randomized search of an estimator.
roc_curve	Draws a receiver operating characteristic (ROC) curve.
stepwise	Uses the stepwise algorithm to find the most suitable number of features when fitting the estimator.
validation_curve	Draws the Validation curve.

verticapy.learn.naive_bayes

Class	Definition
BernoulliNB	i.e. NaiveBayes with param nbtype = 'bernoulli'.
CategoricalNB	i.e. NaiveBayes with param nbtype = 'categorical'.
GaussianNB	i.e. NaiveBayes with param nbtype = 'gaussian'.
MultinomialNB	i.e. NaiveBayes with param nbtype = 'multinomial'.
NaiveBayes	Creates a NaiveBayes object using the Vertica Naive Bayes algorithm.

verticapy.learn.neighbors

Class	Definition
KernelDensity	Creates a KernelDensity object.
KNeighborsClassifier	Creates a KNeighborsClassifier object using the k-nearest neighbors algorithm.
KNeighborsRegressor	Creates a KNeighborsRegressor object using the k-nearest neighbors algorithm.
LocalOutlierFactor	Creates a LocalOutlierFactor object using the Local Outlier Factor algorithm as defined by Markus M. Breunig, Hans-Peter Kriegel, Raymond T. Ng, and Jörg Sander.
NearestCentroid	Creates a NearestCentroid object using the k-nearest centroid algorithm.

verticapy.learn.pipeline

Class	Definition
Pipeline	Creates a Pipeline object, sequentially applying a list of transformations and a final estimator. The intermediate steps must implement a transform method.

verticapy.learn.preprocessing

Class / Function	Definition
Balance	Creates a view with an equal distribution of the input data based on the response_column.
CountVectorizer	Creates a Text Index which will count the occurences of each word in the data.
MinMaxScaler	i.e. Normalizer with param method = 'minmax'.
Normalizer	Creates a Vertica Normalizer object.
OneHotEncoder	Creates a Vertica OneHotEncoder object.
RobustScaler	i.e. Normalizer with param method = 'robust_zscore'.
StandardScaler	i.e. Normalizer with param method = 'zscore'.

verticapy.learn.svm

Class	Definition
LinearSVC	Creates a LinearSVC object using the Vertica SVM algorithm.
LinearSVR	Creates a LinearSVR object using the Vertica SVM algorithm.

verticapy.learn.tree

Class	Definition
DecisionTreeClassifier	Single Decision Tree Classifier.
DecisionTreeRegressor	Single Decision Tree Regressor.
DummyTreeClassifier	A classifier that overfits the training data.
DummyTreeRegressor	A regressor that overfits the training data.

verticapy.learn.tsa

Class	Definition
SARIMAX	Creates an SARIMAX object using the Vertica linear regression algorithm.
VAR	Creates an VAR object using the Vertica linear regression algorithm.

verticapy.learn.utilities

Class	Definition
check_model	Checks if the model already exists.
load_model	Loads a Vertica model and returns the associated object.