Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/*-
*
* * Copyright 2015 Skymind,Inc.
* *
* * Licensed 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.
*
*
*/
package org.nd4j.linalg.indexing;
import com.google.common.base.Function;
import lombok.NonNull;
import org.nd4j.linalg.api.complex.IComplexNDArray;
import org.nd4j.linalg.api.complex.IComplexNumber;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.api.ops.impl.accum.MatchCondition;
import org.nd4j.linalg.api.ops.impl.indexaccum.FirstIndex;
import org.nd4j.linalg.api.ops.impl.indexaccum.LastIndex;
import org.nd4j.linalg.api.ops.impl.transforms.comparison.Choose;
import org.nd4j.linalg.api.ops.impl.transforms.comparison.CompareAndReplace;
import org.nd4j.linalg.api.ops.impl.transforms.comparison.CompareAndSet;
import org.nd4j.linalg.api.shape.Shape;
import org.nd4j.linalg.api.shape.loop.coordinatefunction.CoordinateFunction;
import org.nd4j.linalg.factory.Nd4j;
import org.nd4j.linalg.indexing.conditions.BaseCondition;
import org.nd4j.linalg.indexing.conditions.Condition;
import java.util.List;
import java.util.concurrent.atomic.AtomicBoolean;
/**
* Boolean indexing
*
* @author Adam Gibson
*/
public class BooleanIndexing {
/**
* And
*
* @param n
* @param cond
* @return
*/
public static boolean and(IComplexNDArray n, Condition cond) {
boolean ret = true;
IComplexNDArray linear = n.linearView();
for (int i = 0; i < linear.length(); i++) {
ret = ret && cond.apply(linear.getComplex(i));
}
return ret;
}
/**
* Or over the whole ndarray given some condition
*
* @param n
* @param cond
* @return
*/
public static boolean or(IComplexNDArray n, Condition cond) {
boolean ret = false;
IComplexNDArray linear = n.linearView();
for (int i = 0; i < linear.length(); i++) {
ret = ret || cond.apply(linear.getComplex(i));
}
return ret;
}
/**
* And over the whole ndarray given some condition
*
* @param n the ndarray to test
* @param cond the condition to test against
* @return true if all of the elements meet the specified
* condition false otherwise
*/
public static boolean and(final INDArray n, final Condition cond) {
if (cond instanceof BaseCondition) {
long val = (long) Nd4j.getExecutioner().exec(new MatchCondition(n, cond), Integer.MAX_VALUE).getDouble(0);
if (val == n.lengthLong())
return true;
else
return false;
} else {
boolean ret = true;
final AtomicBoolean a = new AtomicBoolean(ret);
Shape.iterate(n, new CoordinateFunction() {
@Override
public void process(int[]... coord) {
if (a.get())
a.compareAndSet(true, a.get() && cond.apply(n.getDouble(coord[0])));
}
});
return a.get();
}
}
/**
* And over the whole ndarray given some condition, with respect to dimensions
*
* @param n the ndarray to test
* @param condition the condition to test against
* @return true if all of the elements meet the specified
* condition false otherwise
*/
public static boolean[] and(final INDArray n, final Condition condition, int... dimension) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
MatchCondition op = new MatchCondition(n, condition);
INDArray arr = Nd4j.getExecutioner().exec(op, dimension);
boolean[] result = new boolean[arr.length()];
long tadLength = Shape.getTADLength(n.shape(), dimension);
for (int i = 0; i < arr.length(); i++) {
if (arr.getDouble(i) == tadLength)
result[i] = true;
else
result[i] = false;
}
return result;
}
/**
* Or over the whole ndarray given some condition, with respect to dimensions
*
* @param n the ndarray to test
* @param condition the condition to test against
* @return true if all of the elements meet the specified
* condition false otherwise
*/
public static boolean[] or(final INDArray n, final Condition condition, int... dimension) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
MatchCondition op = new MatchCondition(n, condition);
INDArray arr = Nd4j.getExecutioner().exec(op, dimension);
boolean[] result = new boolean[arr.length()];
for (int i = 0; i < arr.length(); i++) {
if (arr.getDouble(i) > 0)
result[i] = true;
else
result[i] = false;
}
return result;
}
/**
* Or over the whole ndarray given some condition
*
* @param n
* @param cond
* @return
*/
public static boolean or(final INDArray n, final Condition cond) {
if (cond instanceof BaseCondition) {
long val = (long) Nd4j.getExecutioner().exec(new MatchCondition(n, cond), Integer.MAX_VALUE).getDouble(0);
if (val > 0)
return true;
else
return false;
} else {
boolean ret = false;
final AtomicBoolean a = new AtomicBoolean(ret);
Shape.iterate(n, new CoordinateFunction() {
@Override
public void process(int[]... coord) {
if (!a.get())
a.compareAndSet(false, a.get() || cond.apply(n.getDouble(coord[0])));
}
});
return a.get();
}
}
/**
* Based on the matching elements
* op to based on condition to with function function
*
* @param to the ndarray to op
* @param condition the condition on op
* @param function the function to apply the op to
*/
public static void applyWhere(final INDArray to, final Condition condition,
final Function function) {
// keep original java implementation for dynamic
Shape.iterate(to, new CoordinateFunction() {
@Override
public void process(int[]... coord) {
if (condition.apply(to.getDouble(coord[0])))
to.putScalar(coord[0], function.apply(to.getDouble(coord[0])).doubleValue());
}
});
}
/**
* This method sets provided number to all elements which match specified condition
*
* @param to
* @param condition
* @param number
*/
public static void applyWhere(final INDArray to, final Condition condition, final Number number) {
if (condition instanceof BaseCondition) {
// for all static conditions we go native
Nd4j.getExecutioner().exec(new CompareAndSet(to, number.doubleValue(), condition));
} else {
final double value = number.doubleValue();
final Function dynamic = new Function() {
@Override
public Number apply(Number number) {
return value;
}
};
Shape.iterate(to, new CoordinateFunction() {
@Override
public void process(int[]... coord) {
if (condition.apply(to.getDouble(coord[0])))
to.putScalar(coord[0], dynamic.apply(to.getDouble(coord[0])).doubleValue());
}
});
}
}
/**
* This method does element-wise comparison
* for 2 equal-sized matrices, for each element that matches Condition.
* To is the array to apply the indexing to
* from is a condition mask array (0 or 1).
* This would come from the output of a bit masking method like:
* {@link INDArray#gt(Number)},{@link INDArray#gte(Number)},
* {@link INDArray#lt(Number)},..
*
* @param to the array to apply the condition to
* @param from the mask array
* @param condition the condition to apply
*/
public static void assignIf(@NonNull INDArray to, @NonNull INDArray from, @NonNull Condition condition) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
if (to.lengthLong() != from.lengthLong())
throw new IllegalStateException("Mis matched length for to and from");
Nd4j.getExecutioner().exec(new CompareAndSet(to, from, condition));
}
/**
* This method does element-wise comparison for 2 equal-sized matrices, for each element that matches Condition
*
* @param to
* @param from
* @param condition
*/
public static void replaceWhere(@NonNull INDArray to, @NonNull INDArray from, @NonNull Condition condition) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
if (to.lengthLong() != from.lengthLong())
throw new IllegalStateException("Mis matched length for to and from");
Nd4j.getExecutioner().exec(new CompareAndReplace(to, from, condition));
}
/**
* Choose from the inputs based on the given condition.
* This returns a row vector of all elements fulfilling the
* condition listed within the array for input
* @param input the input to filter
* @param condition the condition to filter based on
* @return a row vector of the input elements that are true
* ffor the given conditions
*/
public static INDArray chooseFrom(@NonNull INDArray[] input,@NonNull Condition condition) {
Choose choose = new Choose(input,condition);
Nd4j.getExecutioner().exec(choose);
int secondOutput = choose.getOutputArgument(1).getInt(0);
if(secondOutput < 1) {
return null;
}
return choose.getOutputArgument(0);
}
/**
* A minor shortcut for applying a bitmask to
* a matrix
* @param arr The array to apply the mask to
* @param mask the mask to apply
* @return the array with the mask applied
*/
public static INDArray applyMask(INDArray arr,INDArray mask) {
return arr.mul(mask);
}
/**
* A minor shortcut for applying a bitmask to
* a matrix
* @param arr The array to apply the mask to
* @param mask the mask to apply
* @return the array with the mask applied
*/
public static INDArray applyMaskInPlace(INDArray arr,INDArray mask) {
return arr.muli(mask);
}
/**
* Choose from the inputs based on the given condition.
* This returns a row vector of all elements fulfilling the
* condition listed within the array for input.
* The double and integer arguments are only relevant
* for scalar operations (like when you have a scalar
* you are trying to compare each element in your input against)
*
* @param input the input to filter
* @param tArgs the double args
* @param iArgs the integer args
* @param condition the condition to filter based on
* @return a row vector of the input elements that are true
* ffor the given conditions
*/
public static INDArray chooseFrom(@NonNull INDArray[] input, @NonNull List tArgs, @NonNull List iArgs, @NonNull Condition condition) {
Choose choose = new Choose(input,iArgs,tArgs,condition);
Nd4j.getExecutioner().exec(choose);
int secondOutput = choose.getOutputArgument(1).getInt(0);
if(secondOutput < 1) {
return null;
}
INDArray ret = choose.getOutputArgument(0).get(NDArrayIndex.interval(0,secondOutput));
ret = ret.reshape(ret.length());
return ret;
}
/**
* This method does element-wise assessing for 2 equal-sized matrices, for each element that matches Condition
*
* @param to
* @param set
* @param condition
*/
public static void replaceWhere(@NonNull INDArray to, @NonNull Number set, @NonNull Condition condition) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
Nd4j.getExecutioner().exec(new CompareAndSet(to, set.doubleValue(), condition));
}
/**
* Based on the matching elements
* op to based on condition to with function function
*
* @param to the ndarray to op
* @param condition the condition on op
* @param function the function to apply the op to
*/
public static void applyWhere(final INDArray to, final Condition condition, final Function function,
final Function alternativeFunction) {
Shape.iterate(to, new CoordinateFunction() {
@Override
public void process(int[]... coord) {
if (condition.apply(to.getDouble(coord[0]))) {
to.putScalar(coord[0], function.apply(to.getDouble(coord[0])).doubleValue());
} else {
to.putScalar(coord[0], alternativeFunction.apply(to.getDouble(coord[0])).doubleValue());
}
}
});
}
/**
* Based on the matching elements
* op to based on condition to with function function
*
* @param to the ndarray to op
* @param condition the condition on op
* @param function the function to apply the op to
*/
public static void applyWhere(IComplexNDArray to, Condition condition,
Function function) {
IComplexNDArray linear = to.linearView();
for (int i = 0; i < linear.linearView().length(); i++) {
if (condition.apply(linear.getDouble(i))) {
linear.putScalar(i, function.apply(linear.getComplex(i)));
}
}
}
/**
* This method returns first index matching given condition
*
* PLEASE NOTE: This method will return -1 value if condition wasn't met
*
* @param array
* @param condition
* @return
*/
public static INDArray firstIndex(INDArray array, Condition condition) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
FirstIndex idx = new FirstIndex(array, condition);
Nd4j.getExecutioner().exec(idx);
return Nd4j.scalar((double) idx.getFinalResult());
}
/**
* This method returns first index matching given condition along given dimensions
*
* PLEASE NOTE: This method will return -1 values for missing conditions
*
* @param array
* @param condition
* @param dimension
* @return
*/
public static INDArray firstIndex(INDArray array, Condition condition, int... dimension) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
return Nd4j.getExecutioner().exec(new FirstIndex(array, condition), dimension);
}
/**
* This method returns last index matching given condition
*
* PLEASE NOTE: This method will return -1 value if condition wasn't met
*
* @param array
* @param condition
* @return
*/
public static INDArray lastIndex(INDArray array, Condition condition) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
LastIndex idx = new LastIndex(array, condition);
Nd4j.getExecutioner().exec(idx);
return Nd4j.scalar((double) idx.getFinalResult());
}
/**
* This method returns first index matching given condition along given dimensions
*
* PLEASE NOTE: This method will return -1 values for missing conditions
*
* @param array
* @param condition
* @param dimension
* @return
*/
public static INDArray lastIndex(INDArray array, Condition condition, int... dimension) {
if (!(condition instanceof BaseCondition))
throw new UnsupportedOperationException("Only static Conditions are supported");
return Nd4j.getExecutioner().exec(new LastIndex(array, condition), dimension);
}
}
