neural-network/Dataloader.java at main · SparshSuman2005/neural-network · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
import java.io.DataInputStream;
import java.io.FileInputStream;
import java.io.IOException;
import java.util.ArrayList;


public class Dataloader {

     public static ArrayList<Samplecontainer> samples = new ArrayList<>();

     public static void loaddata()throws IOException{
        samples.clear();
         DataInputStream images = new DataInputStream(
            new FileInputStream("archive/train-images-idx3-ubyte/train-images-idx3-ubyte")
        );

        int magic = images.readInt();
        int numberofimages = images.readInt();
        int rows = images.readInt();
        int cols = images.readInt();


        double[][][] x = new double[numberofimages][rows*cols][1];

        System.out.println("Images");


        for(int k = 0 ; k < numberofimages ; k++){
            int index = 0;
            for(int i = 0 ; i < rows ; i++){
            for(int j = 0 ; j < cols ; j++){

                int pixel = images.readUnsignedByte();
                x[k][index][0] = pixel / 255.0;
                index++;
            }
        }
    }
        images.close();

        System.out.println("Labels");


        DataInputStream labels = new DataInputStream(
            new FileInputStream("archive/train-labels-idx1-ubyte/train-labels-idx1-ubyte")
        );

        int labelMagic = labels.readInt();
        int numberOfLabels = labels.readInt();

        if(numberOfLabels!=numberofimages){
            System.out.println("Mismatch dataset!!");
            System.exit(0);
        }


        int[] y_ = new int[numberofimages];

        for(int k = 0 ; k < numberofimages ; k++){
            y_[k]=labels.readUnsignedByte();
        }

        labels.close();

        System.out.println(y_[0]);
        System.out.println(y_[1]);

        System.out.println("One-hot");


        //one-hot encoding
        double[][][] y = new double[numberofimages][10][1];

        for(int i = 0 ; i < numberofimages ; i++){

           int digit = y_[i];

           y[i][digit][0] = 1.0;
        }

        Samplecontainer[] sample = new Samplecontainer[numberofimages];

        System.out.println("container");


        for(int i = 0 ; i < numberofimages ; i++){
            sample[i] = new Samplecontainer(x[i],y[i]);
            samples.add(sample[i]);
        }


     }


     public static void loadtestdata() throws IOException {
        samples.clear();

        DataInputStream images = new DataInputStream(
            new FileInputStream("archive/t10k-images-idx3-ubyte/t10k-images-idx3-ubyte")
        );

        int magic = images.readInt();
        int numberofimages = images.readInt();
        int rows = images.readInt();
        int cols = images.readInt();

        double[][][] x = new double[numberofimages][rows * cols][1];

        System.out.println("Test Images");

        for (int k = 0; k < numberofimages; k++) {
            int index = 0;
            for (int i = 0; i < rows; i++) {
                for (int j = 0; j < cols; j++) {
                    int pixel = images.readUnsignedByte();
                    x[k][index][0] = pixel / 255.0;
                    index++;
                }
            }
        }

        images.close();

        System.out.println("Test Labels");

        DataInputStream labels = new DataInputStream(
            new FileInputStream("archive/t10k-labels-idx1-ubyte/t10k-labels-idx1-ubyte")
        );

        int labelMagic = labels.readInt();
        int numberOfLabels = labels.readInt();

        if (numberOfLabels != numberofimages) {
            System.out.println("Mismatch dataset!!");
            System.exit(0);
        }

        int[] y_ = new int[numberofimages];

        for (int k = 0; k < numberofimages; k++) {
            y_[k] = labels.readUnsignedByte();
        }

        labels.close();

        double[][][] y = new double[numberofimages][10][1];

        for (int i = 0; i < numberofimages; i++) {
            int digit = y_[i];
            y[i][digit][0] = 1.0;
        }

        for (int i = 0; i < numberofimages; i++) {
            samples.add(new Samplecontainer(x[i], y[i]));
        }
    }

    public static void main(String[] args) throws IOException {

        loaddata();


        Matrixutilities.print(samples.get(10).image);
        Matrixutilities.print(samples.get(10).label);


    }
}