decisionTreeRealData2.daphne

#-------------------------------------------------------------
#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
#
# Modifications 2024 The DAPHNE Consortium.
#
#-------------------------------------------------------------

# This script has been manually translated from Apache SystemDS (https://github.com/apache/systemds).
# Original file: src/test/scripts/functions/builtin/decisionTreeRealData2.dml @ 8d4c3feaa84f58e0ce7e305a0e93a1f1bafbd502.

# Test case using the Wine dataset.

import "../../../../scripts/algorithms/decisionTree_.daph";
import "../../../../scripts/algorithms/decisionTreePredict_.daph";
import "../../../../scripts/algorithms/randomForest_.daph";
import "../../../../scripts/algorithms/randomForestPredict_.daph";

F = readFrame($data);

recoded12, dict12 = recode(as.matrix(F[, 12]), false);
# TODO Support n-ary cbind/rbind (see #207).
X = cbind(cbind(cbind(cbind(cbind(cbind(cbind(cbind(cbind(cbind(cbind(cbind(
    bin(as.matrix(F[,  0]), 10),
    bin(as.matrix(F[,  1]), 10)),
    bin(as.matrix(F[,  2]), 10)),
    bin(as.matrix(F[,  3]), 10)),
    bin(as.matrix(F[,  4]), 10)),
    bin(as.matrix(F[,  5]), 10)),
    bin(as.matrix(F[,  6]), 10)),
    bin(as.matrix(F[,  7]), 10)),
    bin(as.matrix(F[,  8]), 10)),
    bin(as.matrix(F[,  9]), 10)),
    bin(as.matrix(F[, 10]), 50)),
    bin(as.matrix(F[, 11]), 10)),
    recoded12
);
# Switch from DAPHNE'S 0-based indexing to SystemDS's 1-based indexing.
X = X + 1;

# TODO as.f64 should not be necessary.
R = reshape(as.f64([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2]), 1, 13);

Y = X[,ncol(X) - 1];
X = X[,0:ncol(X) - 1];
X = replace(X, nan, 5); # 1 val

# TODO Initializing yhat before the if-then-else should not be necessary.
yhat = fill(0.0, 0, 0);
if( $dt==1 ) {
  M = decisionTree_.decisionTree(
    /*X=*/X, /*y=*/Y, /*ctypes=*/R,
    /*max_depth=*/10, /*min_leaf=*/4, /*min_split=*/10,
    /*max_features=*/1.0, /*max_values=*/$maxV, /*max_dataratio=*/0.25,
    /*impurity=*/"gini", /*seed=*/7, /*verbose=*/false
  );
  yhat = decisionTreePredict_.decisionTreePredict(
    /*X=*/X, /*ctypes=*/R, /*M=*/M,
    /*strategy=*/"TT", /*verbose=*/false
  );
}
else {
  sf = 1.0/($dt - 1);
  M = randomForest_.randomForest(
    /*X=*/X, /*y=*/Y, /*ctypes=*/R,
    /*num_trees=*/$dt - 1, /*sample_frac=*/sf,
    /*feature_frac=*/1.0, /*max_depth=*/10, /*min_leaf=*/4, /*min_split=*/10,
    /*max_features=*/1.0, /*max_values=*/$maxV,
    /*impurity=*/"gini", /*seed=*/7, /*verbose=*/false
  );
  yhat = randomForestPredict_.randomForestPredict(
    /*X=*/X, /*y=*/fill(0.0, 0, 0), /*ctypes=*/R, /*M=*/M, /*verbose=*/false
  );
}

acc = mean(yhat == Y);
print(acc);