Nd4jEx1_INDArrayBasics.java

package org.nd4j.examples;

import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.factory.Nd4j;

import java.util.Arrays;

/**
 * --- ND4J Example 1: INDArray Basics ---
 *
 * In this example, we'll see some basic operations for INDArrays
 *
 * @author Alex Black
 */
public class Nd4jEx1_INDArrayBasics {

    public static void main(String[] args) {

        /*
        Before we begin, let's review what an INDArray is:
        A INDArray is a multi-dimensional array of numbers: a vector, matrix, or tensor for example.
        Internally, it may store single precision or double precision floating point values for each entry.

        Here, we'll see how you can get some basic information about INDArrays. In later examples, we'll see
        the different ways to create INDArrays, and more operations we can do on them.
         */

        //Let's start by creating a basic 2d array: a matrix with 3 rows and 5 columns. All elements are 0.0
        int nRows = 3;
        int nColumns = 5;
        INDArray myArray = Nd4j.zeros(nRows, nColumns);

        //Next, print some basic information about the array:
        System.out.println("Basic INDArray information:");
        System.out.println("Num. Rows:          " + myArray.rows());
        System.out.println("Num. Columns:       " + myArray.columns());
        System.out.println("Num. Dimensions:    " + myArray.rank());                    //2 dimensions -> rank 2
        System.out.println("Shape:              " + Arrays.toString(myArray.shape()));  //[3,5] -> 3 rows, 5 columns
        System.out.println("Length:             " + myArray.length());                  // 3 rows * 5 columns = 15 total elements

        //We can print the array itself using toString method:
        System.out.println("\nArray Contents:\n" + myArray);

        //There are some other ways we can get the same or similar info
        System.out.println();
        System.out.println("size(0) == nRows:   " + myArray.size(0));                   //Also equivalent to: .shape()[0]
        System.out.println("size(1) == nCols:   " + myArray.size(1));                   //Also equivalent to: .shape()[1]
        System.out.println("Is a vector:        " + myArray.isVector());
        System.out.println("Is a scalar:        " + myArray.isScalar());
        System.out.println("Is a matrix:        " + myArray.isMatrix());
        System.out.println("Is a square matrix: " + myArray.isSquare());;



        //Let's make some modifications to our array...
        // Note that indexing starts at 0. Thus 0..2 are valid indices for rows, and 0..4 are valid indices for columns here
        myArray.putScalar(0, 1, 2.0);           //Set value at row 0, column 1 to value 2.0
        myArray.putScalar(2, 3, 5.0);           //Set value at row 2, column 3 to value 5.0
        System.out.println("\nArray after putScalar operations:");
        System.out.println(myArray);


        //We can also get individual values:
        double val0 = myArray.getDouble(0, 1);  //Get the value at row 0, column 1 - expect value 2.0 as we set this earlier
        System.out.println("\nValue at (0,1):     " + val0);

        //Finally, there are many things we can do to the array... for example adding scalars:
        INDArray myArray2 = myArray.add(1.0);   //Add 1.0 to each entry
        System.out.println("\nNew INDArray, after adding 1.0 to each entry:");
        System.out.println(myArray2);

        INDArray myArray3 = myArray2.mul(2.0);  //Multiply each entry by 2.0
        System.out.println("\nNew INDArray, after multiplying each entry by 2.0:");
        System.out.println(myArray3);
    }

}