resources/ddl/define-macros.hive

-----------------------------------------------------------------------------
-- Hivemall: Hive scalable Machine Learning Library
-----------------------------------------------------------------------------

-- macros available from hive 0.12.0
-- see https://issues.apache.org/jira/browse/HIVE-2655

--------------------
-- General Macros --
--------------------

create temporary macro java_min(x DOUBLE, y DOUBLE)
reflect("java.lang.Math", "min", x, y);

create temporary macro max2(x DOUBLE, y DOUBLE)
if(x>y,x,y);

create temporary macro min2(x DOUBLE, y DOUBLE)
if(x<y,x,y);

--------------------------
-- Statistics functions --
--------------------------

create temporary macro idf(df_t DOUBLE, n_docs DOUBLE)
log(10, n_docs / max2(1,df_t)) + 1.0;

create temporary macro tfidf(tf FLOAT, df_t DOUBLE, n_docs DOUBLE)
tf * (log(10, n_docs / max2(1,df_t)) + 1.0);

--------------------------
-- Evaluation functions --
--------------------------

-- CAUTION: UDAF in macro is not yet support supported in Hive

-- Root Mean Squared Error
-- create temporary macro RMSE(predicted FLOAT, actual FLOAT)
-- sqrt(sum(pow(predicted - actual,2.0))/count(1));

-- Mean Squared Error
-- create temporary macro MSE(predicted FLOAT, actual FLOAT)
-- sum(pow(predicted - actual,2.0))/count(1);

-- Mean Absolute Error
-- create temporary macro MAE(predicted FLOAT, actual FLOAT)
-- sum(abs(predicted - actual))/count(1);

-- sum of squared errors
-- create temporary macro SSE(predicted FLOAT, actual FLOAT)
-- sum(pow(actual - predicted,2.0));

-- sum of squared total
-- create temporary macro SST(actual FLOAT, mean_actual FLOAT)
-- sum(pow(actual-mean_actual,2.0));

-- coefficient of determination (R^2)
-- create temporary macro R2(predicted FLOAT, actual FLOAT, mean_actual FLOAT)
-- 1 - (SSE(predicted, actual) / SST(actual, mean_actual));