Blog

Platforms: WhereOS, Spark, Hive Class: brickhouse.udf.collect.TruncateArrayUDF More functions can be added to WhereOS via Python

tree_predict(string modelId, string model, array features [, const string options | const boolean classification=false]) –

transpose_and_dot(array X, array Y) – Returns dot(X.T, Y) as array, shape = (X.#cols, Y.#cols) WITH

tree_export(string model, const string options, optional array featureNames=null, optional array classNames=null) – exports a Decision

tree_predict_v1(string modelId, int modelType, string script, array features [, const boolean classification]) – Returns a

train_slim( int i, map r_i, map topKRatesOfI, int j, map r_j [, constant string options])