Implement prepareNeuralNetworkFile

Startcode/neuralNetwork.c

@@ -52,8 +52,8 @@ static int checkFileHeader(FILE *file)
     int fileHeaderLen = strlen(FILE_HEADER_STRING);
     char buffer[BUFFER_SIZE] = {0};
 
-    if(BUFFER_SIZE-1 < fileHeaderLen)
-        fileHeaderLen = BUFFER_SIZE-1;
+    if(BUFFER_SIZE - 1 < fileHeaderLen)
+        fileHeaderLen = BUFFER_SIZE - 1;
 
     if(fread(buffer, sizeof(char), fileHeaderLen, file) == fileHeaderLen)
         isValid = strcmp(buffer, FILE_HEADER_STRING) == 0;
@@ -67,17 +67,17 @@ static unsigned int readDimension(FILE *file)
 
     if(fread(&dimension, sizeof(int), 1, file) != 1)
         dimension = 0;
-    
+
     return dimension;
 }
 
 static Matrix readMatrix(FILE *file, unsigned int rows, unsigned int cols)
 {
     Matrix matrix = createMatrix(rows, cols);
-    
+
     if(matrix.buffer != NULL)
     {
-        if(fread(matrix.buffer, sizeof(MatrixType), rows*cols, file) != rows*cols)
+        if(fread(matrix.buffer, sizeof(MatrixType), rows * cols, file) != rows * cols)
             clearMatrix(&matrix);
     }
 
@@ -110,13 +110,13 @@ static void clearLayer(Layer *layer)
 
 static void assignActivations(NeuralNetwork model)
 {
-    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;
     }
 
     if(model.numberOfLayers > 0)
-        model.layers[model.numberOfLayers-1].activation = softmax;
+        model.layers[model.numberOfLayers - 1].activation = softmax;
 }
 
 NeuralNetwork loadModel(const char *path)
@@ -128,7 +128,7 @@ NeuralNetwork loadModel(const char *path)
     {
         if(checkFileHeader(file))
         {
-            unsigned int inputDimension = readDimension(file); 
+            unsigned int inputDimension = readDimension(file);
             unsigned int outputDimension = readDimension(file);
 
             while(inputDimension > 0 && outputDimension > 0)
@@ -142,7 +142,7 @@ NeuralNetwork loadModel(const char *path)
                     clearModel(&model);
                     break;
                 }
-                
+
                 layerBuffer = (Layer *)realloc(model.layers, (model.numberOfLayers + 1) * sizeof(Layer));
 
                 if(layerBuffer != NULL)
@@ -168,6 +168,59 @@ NeuralNetwork loadModel(const char *path)
     return model;
 }
 
+static void writeFileHeader(FILE *file)
+{
+    fwrite(FILE_HEADER_STRING, sizeof(char), strlen(FILE_HEADER_STRING), file);
+}
+
+static void writeDimension(FILE *file, const unsigned int dimension)
+{
+    fwrite(&dimension, sizeof(unsigned int), 1, file);
+}
+
+static void writeMatrix(FILE *file, const Matrix matrix)
+{
+    fwrite(matrix.buffer, sizeof(MatrixType), matrix.rows * matrix.cols, file);
+}
+
+static void writeLayer(FILE *file, const Layer layer)
+{
+    writeMatrix(file, layer.weights);
+    writeMatrix(file, layer.biases);
+}
+
+void saveModel(const char *path, const NeuralNetwork nn)
+{
+    FILE *file = fopen(path, "wb");
+    if(file == NULL)
+    {
+        return;
+    }
+
+    writeFileHeader(file);
+
+    if(nn.numberOfLayers > 0)
+    {
+        const unsigned int inputDimension = nn.layers[0].weights.cols;
+        writeDimension(file, inputDimension);
+
+        for(unsigned int i = 0; i < nn.numberOfLayers; i++)
+        {
+            const Layer layer = nn.layers[i];
+
+            const unsigned int outputDimension = layer.weights.rows;
+
+            writeDimension(file, outputDimension);
+
+            writeLayer(file, layer);
+        }
+    }
+
+    writeDimension(file, 0);
+
+    fclose(file);
+}
+
 static Matrix imageBatchToMatrixOfImageVectors(const GrayScaleImage images[], unsigned int count)
 {
     Matrix matrix = {NULL, 0, 0};
@@ -201,7 +254,7 @@ static Matrix forward(const NeuralNetwork model, Matrix inputBatch)
         {
             Matrix biasResult;
             Matrix weightResult;
-            
+
             weightResult = multiply(model.layers[i].weights, result);
             clearMatrix(&result);
             biasResult = add(model.layers[i].biases, weightResult);
@@ -248,9 +301,9 @@ unsigned char *predict(const NeuralNetwork model, const GrayScaleImage images[],
     Matrix outputBatch = forward(model, inputBatch);
 
     unsigned char *result = argmax(outputBatch);
-    
+
     clearMatrix(&outputBatch);
-    
+
     return result;
 }
 
@@ -265,4 +318,4 @@ void clearModel(NeuralNetwork *model)
         model->layers = NULL;
         model->numberOfLayers = 0;
     }
-}
+}