ehrnspergersi95041/NeuronalesNetzwerk
1
1
Fork 0
generated from freudenreichan/info2Praktikum-NeuronalesNetz
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Compare commits

base: ehrnspergersi95041:eda98eb343f35f11fc5cd76c5bf767b261e21689
Branches Tags
ehrnspergersi95041:main
ehrnspergersi95041:Simon's-Branch
..
compare: ehrnspergersi95041:e844ca13cb8ef7b3315b2b634e6d81c691e8a66e
Branches Tags
ehrnspergersi95041:Simon's-Branch
ehrnspergersi95041:main

No commits in common. "eda98eb343f35f11fc5cd76c5bf767b261e21689" and "e844ca13cb8ef7b3315b2b634e6d81c691e8a66e" have entirely different histories.
eda98eb343 ... e844ca13cb

4 changed files with 49 additions and 506 deletions
Show Stats Expand all files Collapse all files

181
Start_Windows/imageInput.c
View File

@ -1,183 +1,22 @@
// #include <stdio.h>
// #include <stdlib.h>
// #include <string.h>
// #include "imageInput.h"
//
// #define BUFFER_SIZE 100
// #define FILE_HEADER_STRING "__info2_image_file_format__"
//
// // TODO Implementieren Sie geeignete Hilfsfunktionen für das Lesen der Bildserie aus einer Datei
//
// // TODO Vervollständigen Sie die Funktion readImages unter Benutzung Ihrer Hilfsfunktionen
// GrayScaleImageSeries *readImages(const char *path)
// {
// GrayScaleImageSeries *series = NULL;
//
// return series;
// }
//
// // TODO Vervollständigen Sie die Funktion clearSeries, welche eine Bildserie vollständig aus dem Speicher freigibt
// void clearSeries(GrayScaleImageSeries *series)
// {
//
// }
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "imageInput.h"
#define BUFFER_SIZE 100
#define FILE_HEADER_STRING "__info2_image_file_format__"
// ---------------------------------------------------------
// Hilfsfunktionen (static, nur in diesem Modul sichtbar)
// ---------------------------------------------------------
/**
* Prüft, ob die Datei mit dem korrekten Header-String beginnt.
* Gibt 1 zurück, wenn korrekt, sonst 0.
*/
static int checkFileHeader(FILE *file) {
const char *expectedHeader = FILE_HEADER_STRING;
size_t headerLen = strlen(expectedHeader);
char buffer[100]; // Puffer groß genug für den Header
// Wir lesen genau so viele Bytes, wie der Header lang ist
if (fread(buffer, sizeof(char), headerLen, file) != headerLen) {
return 0; // Lesefehler oder Datei zu kurz
}
// Null-Terminierung sicherstellen für strcmp (obwohl wir auch memcmp nutzen könnten)
buffer[headerLen] = '\0';
if (strcmp(buffer, expectedHeader) != 0) {
return 0; // Header stimmt nicht überein
}
return 1;
}
/**
* Reserviert den Speicher für die Basis-Struktur der Bildserie.
* Reserviert Arrays für 'images' und 'labels', aber noch nicht die Pixel-Buffer der einzelnen Bilder.
*/
static GrayScaleImageSeries *allocateSeries(unsigned short count) {
GrayScaleImageSeries *series = (GrayScaleImageSeries *)malloc(sizeof(GrayScaleImageSeries));
if (series == NULL) return NULL;
series->count = count;
// Speicher für das Array der Bild-Strukturen
series->images = (GrayScaleImage *)calloc(count, sizeof(GrayScaleImage));
// Speicher für das Array der Labels
series->labels = (unsigned char *)calloc(count, sizeof(unsigned char));
if (series->images == NULL || series->labels == NULL) {
// Falls eine Allokation fehlschlägt, alles bisherige freigeben
free(series->images);
free(series->labels);
free(series);
return NULL;
}
// TODO Implementieren Sie geeignete Hilfsfunktionen für das Lesen der Bildserie aus einer Datei
// TODO Vervollständigen Sie die Funktion readImages unter Benutzung Ihrer Hilfsfunktionen
GrayScaleImageSeries *readImages(const char *path)
{
GrayScaleImageSeries *series = NULL;
return series;
}
/**
* Liest ein einzelnes Bild (Pixeldaten) und das zugehörige Label.
*/
static int readSingleImage(FILE *file, GrayScaleImage *image, unsigned char *label, unsigned short width, unsigned short height) {
image->width = width;
image->height = height;
// Speicher für die Pixelwerte reservieren
image->buffer = (GrayScalePixelType *)malloc(width * height * sizeof(GrayScalePixelType));
if (image->buffer == NULL) {
return 0;
}
// Pixelwerte lesen
if (fread(image->buffer, sizeof(GrayScalePixelType), width * height, file) != width * height) {
return 0;
}
// Label lesen
if (fread(label, sizeof(unsigned char), 1, file) != 1) {
return 0;
}
return 1;
}
// ---------------------------------------------------------
// Hauptfunktionen (öffentliche API)
// ---------------------------------------------------------
GrayScaleImageSeries *readImages(const char *path) {
FILE *file = fopen(path, "rb"); // WICHTIG: "rb" für binary mode
if (file == NULL) {
return NULL;
}
// 1. Header prüfen
if (!checkFileHeader(file)) {
fclose(file);
return NULL;
}
// 2. Dimensionen lesen (Anzahl, Breite, Höhe)
unsigned short count, width, height;
int readCount = 0;
readCount += fread(&count, sizeof(unsigned short), 1, file);
readCount += fread(&width, sizeof(unsigned short), 1, file);
readCount += fread(&height, sizeof(unsigned short), 1, file);
if (readCount != 3) {
fclose(file);
return NULL;
}
// 3. Speicherstruktur vorbereiten
GrayScaleImageSeries *series = allocateSeries(count);
if (series == NULL) {
fclose(file);
return NULL;
}
// 4. Jedes Bild einzeln einlesen
for (int i = 0; i < count; i++) {
if (!readSingleImage(file, &series->images[i], &series->labels[i], width, height)) {
// Fehler beim Lesen eines Bildes -> Aufräumen
clearSeries(series);
fclose(file);
return NULL;
}
}
fclose(file);
return series;
}
void clearSeries(GrayScaleImageSeries *series) {
if (series == NULL) return;
// 1. Pixel-Buffer jedes einzelnen Bildes freigeben
if (series->images != NULL) {
for (int i = 0; i < series->count; i++) {
if (series->images[i].buffer != NULL) {
free(series->images[i].buffer);
series->images[i].buffer = NULL;
}
}
// 2. Das Array der Bild-Strukturen freigeben
free(series->images);
}
// 3. Das Label-Array freigeben
if (series->labels != NULL) {
free(series->labels);
}
// 4. Die Hauptstruktur freigeben
free(series);
// TODO Vervollständigen Sie die Funktion clearSeries, welche eine Bildserie vollständig aus dem Speicher freigibt
void clearSeries(GrayScaleImageSeries *series)
{
}

2
Start_Windows/matrix.c
View File

@ -116,7 +116,7 @@ Matrix add(const Matrix matrix1, const Matrix matrix2)
for(unsigned int i = 0; i < resRows; ++i)
{
for(unsigned int j = 0; j < resCols; ++j)
for(unsigned int j = 0; i < resCols; ++j)
{
unsigned int i1 = (matrix1.rows == 1) ? 0 : i;
unsigned int j1 = (matrix1.cols == 1) ? 0 : j;

72
Start_Windows/neuralNetworkTests.c
View File

@ -5,7 +5,24 @@
#include "./unity/unity.h"
#include "neuralNetwork.h"
// --- Implementierung der Hilfsfunktion ---
// TODO
// Das Neuronale Netz
// Inhalte: Strukturen, Dateien schreiben
// Ein neuronales besteht aus verschiedenen Schichten und ihren Parametern.
// Die Struktur NeuralNetwork und die Implementierung in der entsprechenden Quelltextdatei bildet dies ab.
// Für die passenden Unittests fehlt jedoch noch eine Methode, die eine gültige Testdatei erzeugt,
// mit der die Funktionalität getestet werden kann.
// Die Datei beginnt mit dem Identifikationstag __info2_neural_network_file_format__,
// gefolgt von den einzelnen Schichten.
//
// Aufgaben:
// 1) Implementieren Sie die Funktion prepareNeuralNetworkFile() in neuralNetworkTests.c.
// Praktikum Informatik 2
//
// Wintersemester 2025
// 2) Stellen Sie sicher, dass alle Unittests erfolgreich durchlaufen.
// make neuralNetworkTests && ./runNeuralNetworkTests
static void prepareNeuralNetworkFile(const char *path, const NeuralNetwork nn)
{
@ -14,46 +31,44 @@ static void prepareNeuralNetworkFile(const char *path, const NeuralNetwork nn)
return;
}
// 1. Header schreiben (ohne Nullterminator)
// 1. Header schreiben
const char *fileHeader = "__info2_neural_network_file_format__";
fwrite(fileHeader, sizeof(char), strlen(fileHeader), file);
// Prüfen ob Layer existieren
if (nn.numberOfLayers > 0)
{
// 2. Input-Dimension der ersten Schicht schreiben (Spalten der ersten Gewichtsmatrix)
// 2. Die Input-Dimension der ALLERERSTEN Schicht schreiben
// (Das sind die Spalten der Gewichtsmatrix der ersten Schicht)
int inputDim = nn.layers[0].weights.cols;
fwrite(&inputDim, sizeof(int), 1, file);
// 3. Durch alle Schichten iterieren
for(int i = 0; i < nn.numberOfLayers; i++)
{
// Output-Dimension der aktuellen Schicht schreiben (Zeilen der Gewichtsmatrix)
// Die Output-Dimension dieser Schicht schreiben (Zeilen der Gewichtsmatrix)
int outputDim = nn.layers[i].weights.rows;
fwrite(&outputDim, sizeof(int), 1, file);
// 4. Gewichte schreiben
// 4. Gewichte (Weights) schreiben (nur den Buffer, keine Dimensionen mehr!)
// loadModel weiß durch inputDim und outputDim schon, wie groß die Matrix ist.
int weightsCount = nn.layers[i].weights.rows * nn.layers[i].weights.cols;
if (nn.layers[i].weights.buffer != NULL) {
fwrite(nn.layers[i].weights.buffer, sizeof(MatrixType), weightsCount, file);
}
fwrite(nn.layers[i].weights.buffer, sizeof(MatrixType), weightsCount, file);
// 5. Biases schreiben
// 5. Biases schreiben (nur den Buffer)
int biasCount = nn.layers[i].biases.rows * nn.layers[i].biases.cols;
if (nn.layers[i].biases.buffer != NULL) {
fwrite(nn.layers[i].biases.buffer, sizeof(MatrixType), biasCount, file);
}
fwrite(nn.layers[i].biases.buffer, sizeof(MatrixType), biasCount, file);
}
}
// 6. Abbruchsignal (Dimension 0) schreiben
// 6. Eine 0 schreiben, um das Ende der Dimensionen zu signalisieren
// (loadModel bricht die while-Schleife ab, wenn readDimension 0 liefert)
int stopMark = 0;
fwrite(&stopMark, sizeof(int), 1, file);
fclose(file);
}
// --- Unit Tests ---
void test_loadModelReturnsCorrectNumberOfLayers(void)
{
const char *path = "some__nn_test_file.info2";
@ -184,7 +199,9 @@ void test_loadModelFailsOnWrongFileTag(void)
if(file != NULL)
{
const char *fileTag = "info2_neural_network_file_format";
fwrite(fileTag, sizeof(char), strlen(fileTag), file);
fclose(file);
}
@ -222,8 +239,6 @@ void test_clearModelSetsMembersToNull(void)
static void someActivation(Matrix *matrix)
{
if (matrix == NULL || matrix->buffer == NULL) return;
for(int i = 0; i < matrix->rows * matrix->cols; i++)
{
matrix->buffer[i] = fabs(matrix->buffer[i]);
@ -236,37 +251,26 @@ void test_predictReturnsCorrectLabels(void)
GrayScalePixelType imageBuffer1[] = {10, 30, 25, 17};
GrayScalePixelType imageBuffer2[] = {20, 40, 10, 128};
GrayScaleImage inputImages[] = {{.buffer=imageBuffer1, .width=2, .height=2}, {.buffer=imageBuffer2, .width=2, .height=2}};
// Wir nutzen explizite Casts auf MatrixType, um sicherzustellen, dass die Typen stimmen
// (besonders wichtig, falls MatrixType double ist, aber hier Ints stehen)
MatrixType weightsBuffer1[] = {(MatrixType)1, (MatrixType)-2, (MatrixType)3, (MatrixType)-4, (MatrixType)5, (MatrixType)-6, (MatrixType)7, (MatrixType)-8};
MatrixType weightsBuffer2[] = {(MatrixType)-9, (MatrixType)10, (MatrixType)11, (MatrixType)12, (MatrixType)13, (MatrixType)14};
MatrixType weightsBuffer3[] = {(MatrixType)-15, (MatrixType)16, (MatrixType)17, (MatrixType)18, (MatrixType)-19, (MatrixType)20, (MatrixType)21, (MatrixType)22, (MatrixType)23, (MatrixType)-24, (MatrixType)25, (MatrixType)26, (MatrixType)27, (MatrixType)-28, (MatrixType)-29};
MatrixType weightsBuffer1[] = {1, -2, 3, -4, 5, -6, 7, -8};
MatrixType weightsBuffer2[] = {-9, 10, 11, 12, 13, 14};
MatrixType weightsBuffer3[] = {-15, 16, 17, 18, -19, 20, 21, 22, 23, -24, 25, 26, 27, -28, -29};
Matrix weights1 = {.buffer=weightsBuffer1, .rows=2, .cols=4};
Matrix weights2 = {.buffer=weightsBuffer2, .rows=3, .cols=2};
Matrix weights3 = {.buffer=weightsBuffer3, .rows=5, .cols=3};
MatrixType biasBuffer1[] = {(MatrixType)200, (MatrixType)0};
MatrixType biasBuffer2[] = {(MatrixType)0, (MatrixType)-100, (MatrixType)0};
MatrixType biasBuffer3[] = {(MatrixType)0, (MatrixType)-1000, (MatrixType)0, (MatrixType)2000, (MatrixType)0};
MatrixType biasBuffer1[] = {200, 0};
MatrixType biasBuffer2[] = {0, -100, 0};
MatrixType biasBuffer3[] = {0, -1000, 0, 2000, 0};
Matrix biases1 = {.buffer=biasBuffer1, .rows=2, .cols=1};
Matrix biases2 = {.buffer=biasBuffer2, .rows=3, .cols=1};
Matrix biases3 = {.buffer=biasBuffer3, .rows=5, .cols=1};
Layer layers[] = {{.weights=weights1, .biases=biases1, .activation=someActivation}, \
{.weights=weights2, .biases=biases2, .activation=someActivation}, \
{.weights=weights3, .biases=biases3, .activation=someActivation}};
NeuralNetwork netUnderTest = {.layers=layers, .numberOfLayers=3};
unsigned char *predictedLabels = predict(netUnderTest, inputImages, 2);
TEST_ASSERT_NOT_NULL(predictedLabels);
int n = (int)(sizeof(expectedLabels) / sizeof(expectedLabels[0]));
TEST_ASSERT_EQUAL_UINT8_ARRAY(expectedLabels, predictedLabels, n);
free(predictedLabels);
}

300
Start_Windows/neuralnetworktestsold.c
View File

@ -1,300 +0,0 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <math.h>
#include "./unity/unity.h"
#include "neuralNetwork.h"
// TODO
// Das Neuronale Netz
// Inhalte: Strukturen, Dateien schreiben
// Ein neuronales besteht aus verschiedenen Schichten und ihren Parametern.
// Die Struktur NeuralNetwork und die Implementierung in der entsprechenden Quelltextdatei bildet dies ab.
// Für die passenden Unittests fehlt jedoch noch eine Methode, die eine gültige Testdatei erzeugt,
// mit der die Funktionalität getestet werden kann.
// Die Datei beginnt mit dem Identifikationstag __info2_neural_network_file_format__,
// gefolgt von den einzelnen Schichten.
//
// Aufgaben:
// 1) Implementieren Sie die Funktion prepareNeuralNetworkFile() in neuralNetworkTests.c.
// Praktikum Informatik 2
//
// Wintersemester 2025
// 2) Stellen Sie sicher, dass alle Unittests erfolgreich durchlaufen.
// make neuralNetworkTests && ./runNeuralNetworkTests
static void prepareNeuralNetworkFile(const char *path, const NeuralNetwork nn)
{
FILE *file = fopen(path, "wb");
if (file == NULL) {
return;
}
// 1. Header schreiben
const char *fileHeader = "__info2_neural_network_file_format__";
fwrite(fileHeader, sizeof(char), strlen(fileHeader), file);
// Prüfen ob Layer existieren
if (nn.numberOfLayers > 0)
{
// 2. Die Input-Dimension der ALLERERSTEN Schicht schreiben
// (Das sind die Spalten der Gewichtsmatrix der ersten Schicht)
int inputDim = nn.layers[0].weights.cols;
fwrite(&inputDim, sizeof(int), 1, file);
// 3. Durch alle Schichten iterieren
for(int i = 0; i < nn.numberOfLayers; i++)
{
// Die Output-Dimension dieser Schicht schreiben (Zeilen der Gewichtsmatrix)
int outputDim = nn.layers[i].weights.rows;
fwrite(&outputDim, sizeof(int), 1, file);
// 4. Gewichte (Weights) schreiben (nur den Buffer, keine Dimensionen mehr!)
// loadModel weiß durch inputDim und outputDim schon, wie groß die Matrix ist.
int weightsCount = nn.layers[i].weights.rows * nn.layers[i].weights.cols;
fwrite(nn.layers[i].weights.buffer, sizeof(MatrixType), weightsCount, file);
// 5. Biases schreiben (nur den Buffer)
int biasCount = nn.layers[i].biases.rows * nn.layers[i].biases.cols;
fwrite(nn.layers[i].biases.buffer, sizeof(MatrixType), biasCount, file);
}
}
// 6. Eine 0 schreiben, um das Ende der Dimensionen zu signalisieren
// (loadModel bricht die while-Schleife ab, wenn readDimension 0 liefert)
int stopMark = 0;
fwrite(&stopMark, sizeof(int), 1, file);
fclose(file);
}
void test_loadModelReturnsCorrectNumberOfLayers(void)
{
const char *path = "some__nn_test_file.info2";
MatrixType buffer1[] = {1, 2, 3, 4, 5, 6};
MatrixType buffer2[] = {1, 2, 3, 4, 5, 6};
Matrix weights1 = {.buffer=buffer1, .rows=3, .cols=2};
Matrix weights2 = {.buffer=buffer2, .rows=2, .cols=3};
MatrixType buffer3[] = {1, 2, 3};
MatrixType buffer4[] = {1, 2};
Matrix biases1 = {.buffer=buffer3, .rows=3, .cols=1};
Matrix biases2 = {.buffer=buffer4, .rows=2, .cols=1};
Layer layers[] = {{.weights=weights1, .biases=biases1}, {.weights=weights2, .biases=biases2}};
NeuralNetwork expectedNet = {.layers=layers, .numberOfLayers=2};
NeuralNetwork netUnderTest;
prepareNeuralNetworkFile(path, expectedNet);
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_EQUAL_INT(expectedNet.numberOfLayers, netUnderTest.numberOfLayers);
clearModel(&netUnderTest);
}
void test_loadModelReturnsCorrectWeightDimensions(void)
{
const char *path = "some__nn_test_file.info2";
MatrixType weightBuffer[] = {1, 2, 3, 4, 5, 6};
Matrix weights = {.buffer=weightBuffer, .rows=3, .cols=2};
MatrixType biasBuffer[] = {7, 8, 9};
Matrix biases = {.buffer=biasBuffer, .rows=3, .cols=1};
Layer layers[] = {{.weights=weights, .biases=biases}};
NeuralNetwork expectedNet = {.layers=layers, .numberOfLayers=1};
NeuralNetwork netUnderTest;
prepareNeuralNetworkFile(path, expectedNet);
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_TRUE(netUnderTest.numberOfLayers > 0);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].weights.rows, netUnderTest.layers[0].weights.rows);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].weights.cols, netUnderTest.layers[0].weights.cols);
clearModel(&netUnderTest);
}
void test_loadModelReturnsCorrectBiasDimensions(void)
{
const char *path = "some__nn_test_file.info2";
MatrixType weightBuffer[] = {1, 2, 3, 4, 5, 6};
Matrix weights = {.buffer=weightBuffer, .rows=3, .cols=2};
MatrixType biasBuffer[] = {7, 8, 9};
Matrix biases = {.buffer=biasBuffer, .rows=3, .cols=1};
Layer layers[] = {{.weights=weights, .biases=biases}};
NeuralNetwork expectedNet = {.layers=layers, .numberOfLayers=1};
NeuralNetwork netUnderTest;
prepareNeuralNetworkFile(path, expectedNet);
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_TRUE(netUnderTest.numberOfLayers > 0);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].biases.rows, netUnderTest.layers[0].biases.rows);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].biases.cols, netUnderTest.layers[0].biases.cols);
clearModel(&netUnderTest);
}
void test_loadModelReturnsCorrectWeights(void)
{
const char *path = "some__nn_test_file.info2";
MatrixType weightBuffer[] = {1, 2, 3, 4, 5, 6};
Matrix weights = {.buffer=weightBuffer, .rows=3, .cols=2};
MatrixType biasBuffer[] = {7, 8, 9};
Matrix biases = {.buffer=biasBuffer, .rows=3, .cols=1};
Layer layers[] = {{.weights=weights, .biases=biases}};
NeuralNetwork expectedNet = {.layers=layers, .numberOfLayers=1};
NeuralNetwork netUnderTest;
prepareNeuralNetworkFile(path, expectedNet);
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_TRUE(netUnderTest.numberOfLayers > 0);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].weights.rows, netUnderTest.layers[0].weights.rows);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].weights.cols, netUnderTest.layers[0].weights.cols);
int n = netUnderTest.layers[0].weights.rows * netUnderTest.layers[0].weights.cols;
TEST_ASSERT_EQUAL_INT_ARRAY(expectedNet.layers[0].weights.buffer, netUnderTest.layers[0].weights.buffer, n);
clearModel(&netUnderTest);
}
void test_loadModelReturnsCorrectBiases(void)
{
const char *path = "some__nn_test_file.info2";
MatrixType weightBuffer[] = {1, 2, 3, 4, 5, 6};
Matrix weights = {.buffer=weightBuffer, .rows=3, .cols=2};
MatrixType biasBuffer[] = {7, 8, 9};
Matrix biases = {.buffer=biasBuffer, .rows=3, .cols=1};
Layer layers[] = {{.weights=weights, .biases=biases}};
NeuralNetwork expectedNet = {.layers=layers, .numberOfLayers=1};
NeuralNetwork netUnderTest;
prepareNeuralNetworkFile(path, expectedNet);
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_TRUE(netUnderTest.numberOfLayers > 0);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].weights.rows, netUnderTest.layers[0].weights.rows);
TEST_ASSERT_EQUAL_INT(expectedNet.layers[0].weights.cols, netUnderTest.layers[0].weights.cols);
int n = netUnderTest.layers[0].biases.rows * netUnderTest.layers[0].biases.cols;
TEST_ASSERT_EQUAL_INT_ARRAY(expectedNet.layers[0].biases.buffer, netUnderTest.layers[0].biases.buffer, n);
clearModel(&netUnderTest);
}
void test_loadModelFailsOnWrongFileTag(void)
{
const char *path = "some_nn_test_file.info2";
NeuralNetwork netUnderTest;
FILE *file = fopen(path, "wb");
if(file != NULL)
{
const char *fileTag = "info2_neural_network_file_format";
fwrite(fileTag, sizeof(char), strlen(fileTag), file);
fclose(file);
}
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_NULL(netUnderTest.layers);
TEST_ASSERT_EQUAL_INT(0, netUnderTest.numberOfLayers);
}
void test_clearModelSetsMembersToNull(void)
{
const char *path = "some__nn_test_file.info2";
MatrixType weightBuffer[] = {1, 2, 3, 4, 5, 6};
Matrix weights = {.buffer=weightBuffer, .rows=3, .cols=2};
MatrixType biasBuffer[] = {7, 8, 9};
Matrix biases = {.buffer=biasBuffer, .rows=3, .cols=1};
Layer layers[] = {{.weights=weights, .biases=biases}};
NeuralNetwork expectedNet = {.layers=layers, .numberOfLayers=1};
NeuralNetwork netUnderTest;
prepareNeuralNetworkFile(path, expectedNet);
netUnderTest = loadModel(path);
remove(path);
TEST_ASSERT_NOT_NULL(netUnderTest.layers);
TEST_ASSERT_TRUE(netUnderTest.numberOfLayers > 0);
clearModel(&netUnderTest);
TEST_ASSERT_NULL(netUnderTest.layers);
TEST_ASSERT_EQUAL_INT(0, netUnderTest.numberOfLayers);
}
static void someActivation(Matrix *matrix)
{
for(int i = 0; i < matrix->rows * matrix->cols; i++)
{
matrix->buffer[i] = fabs(matrix->buffer[i]);
}
}
void test_predictReturnsCorrectLabels(void)
{
const unsigned char expectedLabels[] = {4, 2};
GrayScalePixelType imageBuffer1[] = {10, 30, 25, 17};
GrayScalePixelType imageBuffer2[] = {20, 40, 10, 128};
GrayScaleImage inputImages[] = {{.buffer=imageBuffer1, .width=2, .height=2}, {.buffer=imageBuffer2, .width=2, .height=2}};
MatrixType weightsBuffer1[] = {1, -2, 3, -4, 5, -6, 7, -8};
MatrixType weightsBuffer2[] = {-9, 10, 11, 12, 13, 14};
MatrixType weightsBuffer3[] = {-15, 16, 17, 18, -19, 20, 21, 22, 23, -24, 25, 26, 27, -28, -29};
Matrix weights1 = {.buffer=weightsBuffer1, .rows=2, .cols=4};
Matrix weights2 = {.buffer=weightsBuffer2, .rows=3, .cols=2};
Matrix weights3 = {.buffer=weightsBuffer3, .rows=5, .cols=3};
MatrixType biasBuffer1[] = {200, 0};
MatrixType biasBuffer2[] = {0, -100, 0};
MatrixType biasBuffer3[] = {0, -1000, 0, 2000, 0};
Matrix biases1 = {.buffer=biasBuffer1, .rows=2, .cols=1};
Matrix biases2 = {.buffer=biasBuffer2, .rows=3, .cols=1};
Matrix biases3 = {.buffer=biasBuffer3, .rows=5, .cols=1};
Layer layers[] = {{.weights=weights1, .biases=biases1, .activation=someActivation}, \
{.weights=weights2, .biases=biases2, .activation=someActivation}, \
{.weights=weights3, .biases=biases3, .activation=someActivation}};
NeuralNetwork netUnderTest = {.layers=layers, .numberOfLayers=3};
unsigned char *predictedLabels = predict(netUnderTest, inputImages, 2);
TEST_ASSERT_NOT_NULL(predictedLabels);
int n = (int)(sizeof(expectedLabels) / sizeof(expectedLabels[0]));
TEST_ASSERT_EQUAL_UINT8_ARRAY(expectedLabels, predictedLabels, n);
free(predictedLabels);
}
void setUp(void) {
// Falls notwendig, kann hier Vorbereitungsarbeit gemacht werden
}
void tearDown(void) {
// Hier kann Bereinigungsarbeit nach jedem Test durchgeführt werden
}
int main()
{
UNITY_BEGIN();
printf("\n============================\nNeural network tests\n============================\n");
RUN_TEST(test_loadModelReturnsCorrectNumberOfLayers);
RUN_TEST(test_loadModelReturnsCorrectWeightDimensions);
RUN_TEST(test_loadModelReturnsCorrectBiasDimensions);
RUN_TEST(test_loadModelReturnsCorrectWeights);
RUN_TEST(test_loadModelReturnsCorrectBiases);
RUN_TEST(test_loadModelFailsOnWrongFileTag);
RUN_TEST(test_clearModelSetsMembersToNull);
RUN_TEST(test_predictReturnsCorrectLabels);
return UNITY_END();
}