So meine Freunde jetzt passt das aber auch

neuralNetwork.c

@@ -2,24 +2,24 @@
 #include <stdio.h>
 #include <math.h>
 #include <string.h>
-#include <stdint.h>
 #include "neuralNetwork.h"
 
-#define BUFFER_SIZE 200
+#define BUFFER_SIZE 100
 #define FILE_HEADER_STRING "__info2_neural_network_file_format__"
 
 static void softmax(Matrix *matrix)
 {
     if(matrix->cols > 0)
     {
-        double *colSums = (double *)calloc((size_t)matrix->cols, sizeof(double));
+        double *colSums = (double *)calloc(matrix->cols, sizeof(double));
-        if(colSums == NULL) return;
 
+        if(colSums != NULL)
+        {
             for(int colIdx = 0; colIdx < matrix->cols; colIdx++)
             {
                 for(int rowIdx = 0; rowIdx < matrix->rows; rowIdx++)
                 {
-                MatrixType expValue = (MatrixType)exp(getMatrixAt(*matrix, rowIdx, colIdx));
+                    MatrixType expValue = exp(getMatrixAt(*matrix, rowIdx, colIdx));
                     setMatrixAt(expValue, *matrix, rowIdx, colIdx);
                     colSums[colIdx] += expValue;
                 }
@@ -27,17 +27,15 @@ static void softmax(Matrix *matrix)
 
             for(int colIdx = 0; colIdx < matrix->cols; colIdx++)
             {
-            double s = colSums[colIdx];
-            if(s == 0.0) s = 1.0;
                 for(int rowIdx = 0; rowIdx < matrix->rows; rowIdx++)
                 {
-                MatrixType normalizedValue = (MatrixType)(getMatrixAt(*matrix, rowIdx, colIdx) / s);
+                    MatrixType normalizedValue = getMatrixAt(*matrix, rowIdx, colIdx) / colSums[colIdx];
                     setMatrixAt(normalizedValue, *matrix, rowIdx, colIdx);
                 }
             }
-
             free(colSums);
         }
+    }
 }
 
 static void relu(Matrix *matrix)
@@ -48,54 +46,40 @@ static void relu(Matrix *matrix)
     }
 }
 
-/* Prüft den Dateikopf. Liefert 1 bei Erfolg, 0 bei Fehler. */
 static int checkFileHeader(FILE *file)
 {
-    if(file == NULL) return 0;
+    int isValid = 0;
+    int fileHeaderLen = strlen(FILE_HEADER_STRING);
+    char buffer[BUFFER_SIZE] = {0};
 
-    size_t headerLen = strlen(FILE_HEADER_STRING);
+    if(BUFFER_SIZE-1 < fileHeaderLen)
-    if(headerLen == 0 || headerLen >= BUFFER_SIZE) return 0;
+        fileHeaderLen = BUFFER_SIZE-1;
 
-    char buffer[BUFFER_SIZE];
+    if(fread(buffer, sizeof(char), fileHeaderLen, file) == fileHeaderLen)
-    if(fseek(file, 0, SEEK_SET) != 0) return 0;
+        isValid = strcmp(buffer, FILE_HEADER_STRING) == 0;
-    if(fread(buffer, sizeof(char), headerLen, file) != headerLen) return 0;
-    if(memcmp(buffer, FILE_HEADER_STRING, headerLen) != 0) return 0;
 
-    return 1;
+    return isValid;
 }
 
-/* Liest eine Dimension (wie vom Test-Writer mit sizeof(int) geschrieben) und prüft Plausibilität.
-   Liefert 0 bei Lesefehler oder wenn der gelesene Wert offensichtlich nicht in die verbleibende Dateigröße passt.
-*/
 static unsigned int readDimension(FILE *file)
 {
-    if (file == NULL) return 0;
+    int dimension = 0;
 
-    int value = 0;
+    if(fread(&dimension, sizeof(int), 1, file) != 1)
-    if (fread(&value, sizeof(int), 1, file) != 1) {
+        dimension = 0;
-        return 0;
-    }
 
-    if (value < 0) return 0;
+    return dimension;
-    return (unsigned int)value;
 }
 
-
-/* Liest eine Matrix rows x cols; wenn Einlese-Fehler: leere Matrix zurückgeben */
 static Matrix readMatrix(FILE *file, unsigned int rows, unsigned int cols)
 {
     Matrix matrix = createMatrix(rows, cols);
-    if(matrix.buffer == NULL) return matrix;
 
-    size_t toRead = (size_t)rows * cols;
+    if(matrix.buffer != NULL)
-    if(toRead > 0)
-    {
-        size_t readCount = fread(matrix.buffer, sizeof(MatrixType), toRead, file);
-        if(readCount != toRead)
     {
+        if(fread(matrix.buffer, sizeof(MatrixType), rows*cols, file) != rows*cols)
             clearMatrix(&matrix);
     }
-    }
 
     return matrix;
 }
@@ -103,113 +87,91 @@ static Matrix readMatrix(FILE *file, unsigned int rows, unsigned int cols)
 static Layer readLayer(FILE *file, unsigned int inputDimension, unsigned int outputDimension)
 {
     Layer layer;
-    layer.activation = NULL;
     layer.weights = readMatrix(file, outputDimension, inputDimension);
     layer.biases = readMatrix(file, outputDimension, 1);
+
     return layer;
 }
 
 static int isEmptyLayer(const Layer layer)
 {
-    return layer.biases.cols == 0 || layer.biases.rows == 0 || layer.biases.buffer == NULL ||
+    return layer.biases.cols == 0 || layer.biases.rows == 0 || layer.biases.buffer == NULL || layer.weights.rows == 0 || layer.weights.cols == 0 || layer.weights.buffer == NULL;
-           layer.weights.rows == 0 || layer.weights.cols == 0 || layer.weights.buffer == NULL;
 }
 
 static void clearLayer(Layer *layer)
 {
-    if(layer == NULL) return;
+    if(layer != NULL)
+    {
         clearMatrix(&layer->weights);
         clearMatrix(&layer->biases);
         layer->activation = NULL;
+    }
 }
 
 static void assignActivations(NeuralNetwork model)
 {
-    if(model.numberOfLayers == 0) return;
+    for(int i = 0; i < (int)model.numberOfLayers-1; i++)
-    for(int i = 0; i < (int)model.numberOfLayers - 1; i++)
+    {
         model.layers[i].activation = relu;
-    model.layers[model.numberOfLayers - 1].activation = softmax;
+    }
+
+    if(model.numberOfLayers > 0)
+        model.layers[model.numberOfLayers-1].activation = softmax;
 }
 
 NeuralNetwork loadModel(const char *path)
 {
     NeuralNetwork model = {NULL, 0};
     FILE *file = fopen(path, "rb");
-    if(file == NULL) return model;
 
-    if(fseek(file, 0, SEEK_SET) != 0)
+    if(file != NULL)
     {
-        fclose(file);
+        if(checkFileHeader(file))
-        return model;
-    }
-
-    if(!checkFileHeader(file))
         {
-        fclose(file);
-        return model;
-    }
-
-    /* Lese erstes Dimensions-Paar (input, output) */
             unsigned int inputDimension = readDimension(file);
             unsigned int outputDimension = readDimension(file);
 
-    //fprintf(stderr, "[loadModel] first dims: input=%u output=%u\n", inputDimension, outputDimension);
+            while(inputDimension > 0 && outputDimension > 0)
-
-    while (inputDimension > 0 && outputDimension > 0)
             {
                 Layer layer = readLayer(file, inputDimension, outputDimension);
+                Layer *layerBuffer = NULL;
 
-        if (isEmptyLayer(layer))
+                if(isEmptyLayer(layer))
                 {
                     clearLayer(&layer);
                     clearModel(&model);
                     break;
                 }
 
-        Layer *tmp = (Layer *)realloc(model.layers, (model.numberOfLayers + 1) * sizeof(Layer));
+                layerBuffer = (Layer *)realloc(model.layers, (model.numberOfLayers + 1) * sizeof(Layer));
-        if (tmp == NULL)
+
+                if(layerBuffer != NULL)
+                    model.layers = layerBuffer;
+                else
                 {
-            clearLayer(&layer);
                     clearModel(&model);
                     break;
                 }
-        model.layers = tmp;
+
                 model.layers[model.numberOfLayers] = layer;
                 model.numberOfLayers++;
 
-        /* fprintf(stderr, "[loadModel] loaded layer %d: weights %u x %u, biases %u x %u\n",
+                inputDimension = outputDimension;
-                model.numberOfLayers,
+                outputDimension = readDimension(file);
-                layer.weights.rows, layer.weights.cols,
-                layer.biases.rows, layer.biases.cols); */
-
-        /* Lese das nächste Dimensions-Paar (writer schreibt für jede Schicht ein Paar) */
-        unsigned int nextInput = readDimension(file);
-        unsigned int nextOutput = readDimension(file);
-
-        //fprintf(stderr, "[loadModel] next raw dims read: nextInput=%u nextOutput=%u\n", nextInput, nextOutput);
-
-        /* Wenn nächstes Paar (0,0) -> Ende */
-        if (nextInput == 0 || nextOutput == 0)
-        {
-            inputDimension = 0;
-            outputDimension = 0;
-            break;
             }
-
-        /* Setze für die nächste Iteration */
-        inputDimension = nextInput;
-        outputDimension = nextOutput;
-
-        //fprintf(stderr, "[loadModel] next dims: input=%u output=%u\n", inputDimension, outputDimension);
         }
-
         fclose(file);
+
         assignActivations(model);
+    }
+
     return model;
 }
 
 static Matrix imageBatchToMatrixOfImageVectors(const GrayScaleImage images[], unsigned int count)
 {
+       //Matrix matrix = {NULL, 0, 0};
+    // Explizite Initialisierung verwenden, um die Feldreihenfolge in matrix.h zu umgehen:
     Matrix matrix;
     matrix.buffer = NULL;
     matrix.rows = 0;
@@ -218,15 +180,18 @@ static Matrix imageBatchToMatrixOfImageVectors(const GrayScaleImage images[], un
     if(count > 0 && images != NULL)
     {
         matrix = createMatrix(images[0].height * images[0].width, count);
+
         if(matrix.buffer != NULL)
         {
-            for(unsigned int i = 0; i < count; i++)
+            for(int i = 0; i < count; i++)
+            {
+                for(int j = 0; j < images[i].width * images[i].height; j++)
                 {
-                for(unsigned int j = 0; j < images[i].width * images[i].height; j++)
                     setMatrixAt((MatrixType)images[i].buffer[j], matrix, j, i);
                 }
             }
         }
+    }
 
     return matrix;
 }
@@ -234,42 +199,51 @@ static Matrix imageBatchToMatrixOfImageVectors(const GrayScaleImage images[], un
 static Matrix forward(const NeuralNetwork model, Matrix inputBatch)
 {
     Matrix result = inputBatch;
-    if(result.buffer == NULL) return result;
 
+    if(result.buffer != NULL)
+    {
         for(int i = 0; i < model.numberOfLayers; i++)
         {
-        Matrix weightResult = multiply(model.layers[i].weights, result);
+            Matrix biasResult;
-        clearMatrix(&result);
+            Matrix weightResult;
 
-        Matrix biasResult = add(weightResult, model.layers[i].biases);
+            weightResult = multiply(model.layers[i].weights, result);
+            clearMatrix(&result);
+            biasResult = add(model.layers[i].biases, weightResult);
             clearMatrix(&weightResult);
 
             if(model.layers[i].activation != NULL)
                 model.layers[i].activation(&biasResult);
-
             result = biasResult;
         }
+    }
 
     return result;
 }
 
 unsigned char *argmax(const Matrix matrix)
 {
-    if(matrix.rows == 0 || matrix.cols == 0) return NULL;
+    unsigned char *maxIdx = NULL;
 
-    unsigned char *maxIdx = (unsigned char *)malloc((size_t)matrix.cols * sizeof(unsigned char));
+    if(matrix.rows > 0 && matrix.cols > 0)
-    if(maxIdx == NULL) return NULL;
+    {
+        maxIdx = (unsigned char *)malloc(sizeof(unsigned char) * matrix.cols);
 
+        if(maxIdx != NULL)
+        {
             for(int colIdx = 0; colIdx < matrix.cols; colIdx++)
             {
-        int best = 0;
+                maxIdx[colIdx] = 0;
+
                 for(int rowIdx = 1; rowIdx < matrix.rows; rowIdx++)
                 {
-            if(getMatrixAt(matrix, rowIdx, colIdx) > getMatrixAt(matrix, best, colIdx))
+                    if(getMatrixAt(matrix, rowIdx, colIdx) > getMatrixAt(matrix, maxIdx[colIdx], colIdx))
-                best = rowIdx;
+                        maxIdx[colIdx] = rowIdx;
                 }
-        maxIdx[colIdx] = (unsigned char)best;
             }
+        }
+    }
+
     return maxIdx;
 }
 
@@ -277,17 +251,23 @@ unsigned char *predict(const NeuralNetwork model, const GrayScaleImage images[],
 {
     Matrix inputBatch = imageBatchToMatrixOfImageVectors(images, numberOfImages);
     Matrix outputBatch = forward(model, inputBatch);
+
     unsigned char *result = argmax(outputBatch);
+
     clearMatrix(&outputBatch);
+
     return result;
 }
 
 void clearModel(NeuralNetwork *model)
 {
-    if(model == NULL) return;
+    if(model != NULL)
+    {
         for(int i = 0; i < model->numberOfLayers; i++)
+        {
             clearLayer(&model->layers[i]);
-    free(model->layers);
+        }
         model->layers = NULL;
         model->numberOfLayers = 0;
+    }
 }

neuralNetworkTests.c

@@ -9,48 +9,51 @@
 
 static void prepareNeuralNetworkFile(const char *path, const NeuralNetwork nn)
 {
-    // TODO
     FILE *file = fopen(path, "wb");
     if (!file) return;
 
     // 1) Header-Tag WORTGENAU (OHNE Nullterminator) schreiben
     fwrite(FILE_HEADER_STRING, sizeof(char), strlen(FILE_HEADER_STRING), file);
 
-    // 2) Für jeden Layer die Paare (inputDim, outputDim) und dann Gewichte & Biases schreiben
+    // 2) Layer-Daten im Format, das loadModel() erwartet
-    // inputDim == weights.cols
-    // outputDim == weights.rows
     for (unsigned int i = 0; i < nn.numberOfLayers; ++i)
     {
         const Layer *lay = &nn.layers[i];
 
-        unsigned int inputDim  = lay->weights.cols;
+        int inputDim  = (int)lay->weights.cols;   // cols == inputDimension
-        unsigned int outputDim = lay->weights.rows;
+        int outputDim = (int)lay->weights.rows;   // rows == outputDimension
 
-        // schreibe Dimensionen (4-Byte int / unsigned int)
+        if (i == 0) {
-        fwrite(&inputDim,  sizeof(unsigned int), 1, file);
+            // Erstes Paar: input und output schreiben
-        fwrite(&outputDim, sizeof(unsigned int), 1, file);
+            fwrite(&inputDim,  sizeof(int), 1, file);
+            fwrite(&outputDim, sizeof(int), 1, file);
+        } else {
+            // Ab dem zweiten Layer: NUR das neue outputDimension schreiben
+            fwrite(&outputDim, sizeof(int), 1, file);
+        }
 
-        // schreibe Gewichtsmatrix (row-major: rows*cols Elemente vom Typ MatrixType)
+        // Gewichtsmatrix (row-major)
         size_t weightCount = (size_t)lay->weights.rows * (size_t)lay->weights.cols;
         if (weightCount > 0 && lay->weights.buffer != NULL) {
             fwrite(lay->weights.buffer, sizeof(MatrixType), weightCount, file);
         }
 
-        // schreibe Biasmatrix (rows * cols) - normalerweise rows x 1
+        // Biases (rows x 1)
         size_t biasCount = (size_t)lay->biases.rows * (size_t)lay->biases.cols;
         if (biasCount > 0 && lay->biases.buffer != NULL) {
             fwrite(lay->biases.buffer, sizeof(MatrixType), biasCount, file);
         }
     }
 
-    // 3) Endmarkierung: nächstes Input / Output = 0, damit loadModel() die Schleife beendet
+    // 3) Endmarkierung: EINE 0 (als int) schreiben
-    unsigned int zero = 0;
+    int zero = 0;
-    fwrite(&zero, sizeof(unsigned int), 1, file);
+    fwrite(&zero, sizeof(int), 1, file);
-    fwrite(&zero, sizeof(unsigned int), 1, file);
 
     fclose(file);
 }
 
+
+
 void test_loadModelReturnsCorrectNumberOfLayers(void)
 {
     const char *path = "some__nn_test_file.info2";