scripts/datagen/genRandData4MultiClassSVM.dml

#-------------------------------------------------------------
#
# 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.
#
#-------------------------------------------------------------

# generates random data to test linear logistic regression

# $1 is number of samples
# $2 is number of features (independent variables)
# $3 is maximum feature value (absolute value)
# $4 is maximum weight (absolute value)
# $5 is location to store generated weights
# $6 is location to store generated data
# $7 is location to store generated labels
# $8 addNoise. if 0 then no noise is added, to add noise set this to 1
# $9 is b, 0 disables intercept
# $10 controls sparsity in the generated data

numSamples = $1
numFeatures = $2
maxFeatureValue = $3
maxWeight = $4
addNoise = $8
b = $9

X = Rand(rows=numSamples, cols=numFeatures, min=-1, max=1, pdf="uniform", seed=0, sparsity=$10)
X = X * maxFeatureValue 

w = Rand(rows=numFeatures, cols=1, min=-1, max=1, pdf="uniform", seed=0)
w = w * maxWeight

ot = X%*%w
if(b!=0) {
	b_mat = Rand(rows=1, cols=1, min=b, max=b, pdf="uniform")
	w =  t(append(t(w), b_mat))
	ot = ot + b
}

prob = 1/(1+exp(-ot))
if(addNoise == 1){
	r = Rand(rows=numSamples, cols=1, min=0, max=1, pdf="uniform", seed=0)
}else{
	print("this data generator generates the same dataset for both noise=0 and noise=1")
	r = Rand(rows=numSamples, cols=1, min=0, max=1, pdf="uniform", seed=0)
	#r = Rand(rows=numSamples, cols=1, min=0.5, max=0.5, pdf="uniform")
}
Y = 1 - 2 * (prob < r)
Y = (Y+3)/2

write(w, $5, format="binary")
write(X, $6, format="binary")
write(Y, $7, format="binary")