matrix.c aenderungen

.gitignore

@@ -2,3 +2,9 @@ mnist
 runTests
 *.o
 *.exe
+.vscode/settings.json
+.vscode/launch.json
+.vscode/settings.json
+.vscode/settings.json
+runImageInputTests
+testFile.info2

imageInput.c

@@ -4,6 +4,8 @@
 #include <string.h>
 
 #define FILE_HEADER_STRING "__info2_image_file_format__"
+// define BUFFER 100
+// 10x10 pixel
 
 /* ----------------------------------------------------------
    1. Header prüfen
@@ -28,6 +30,7 @@ static int readMeta(FILE *file, unsigned short *count, unsigned short *width,
     return 0;
   if (fread(height, sizeof(unsigned short), 1, file) != 1)
     return 0;
+
   return 1;
 }
 
@@ -39,14 +42,14 @@ static int readSingleImage(FILE *file, GrayScaleImage *img,
   img->width = width;
   img->height = height;
 
-  size_t numPixels = (size_t)width * (size_t)height;
+  size_t numPixels = (size_t)width * (size_t)height; // anzahl an pixeln
   img->buffer = malloc(numPixels);
   if (!img->buffer)
     return 0;
 
   if (fread(img->buffer, 1, numPixels, file) != numPixels) {
     free(img->buffer);
-    img->buffer = NULL;
+    img->buffer = NULL; // fehler bei ungültiger eingabe
     return 0;
   }
   return 1;
@@ -74,9 +77,11 @@ GrayScaleImageSeries *readImages(const char *path) {
 
   unsigned short count, width, height;
   if (!readMeta(file, &count, &width, &height)) {
+
     fclose(file);
     return NULL;
   }
+  // printf("%d, %d, %d", count, width, height);
 
   GrayScaleImageSeries *series = malloc(sizeof(GrayScaleImageSeries));
   if (!series) {

imageInputTests.c

@@ -126,19 +126,25 @@ void test_readImagesFailsOnWrongFileTag(void) {
   remove(path);
 }
 
-void test_read_GrayScale_Pixel(void) {
-  GrayScaleImageSeries *series = NULL;
+// Test
+
+void test_read_GrayScale_Pixel(
+    void) { // testet das einlesen eines graustufenbildes von readImages()
+  GrayScaleImageSeries *series = NULL; // enthält später das Bild
   const char *path = "testFile.info2";
 
-  prepareImageFile(path, 8, 8, 1, 1);
+  prepareImageFile(path, 8, 8, 1,
+                   1); // Höhe x Breite in Pixel, Anzahl Bilder und Kategorie
   series = readImages(path);
 
-  TEST_ASSERT_NOT_NULL(series);
-  TEST_ASSERT_NOT_NULL(series->images);
-  TEST_ASSERT_EQUAL_UINT(1, series->count);
+  TEST_ASSERT_NOT_NULL(series);             // Speicher reservieren
+  TEST_ASSERT_NOT_NULL(series->images);     // Inhalt ist da
+  TEST_ASSERT_EQUAL_UINT(1, series->count); // Anzahl der Bilder stimmt
 
   for (int i = 0; i < (8 * 8); i++) {
-    TEST_ASSERT_EQUAL_UINT8((GrayScalePixelType)i, series->images[0].buffer[i]);
+    TEST_ASSERT_EQUAL_UINT8(
+        (GrayScalePixelType)i,
+        series->images[0].buffer[i]); // alle Pixelwerte prüfen
   }
 
   clearSeries(series);

makefile

@@ -59,7 +59,8 @@ imageInputTests: imageInput.o imageInputTests.c $(unityfolder)/unity.c
 # --------------------------
 clean:
 ifeq ($(OS),Windows_NT)
-	del /f *.o *.exe
-else
 	rm -f *.o mnist runMatrixTests runNeuralNetworkTests runImageInputTests
+else
+	del /f *.o *.exe
 endif
+ 

matrix.c

@@ -3,35 +3,22 @@
 #include <stdlib.h>
 #include <string.h>
 
-// TODO Matrix-Funktionen implementieren
 /*typedef struct {
     unsigned int rows;  //Zeilen
     unsigned int cols; //Spalten
     MatrixType *buffer; //Zeiger auf Speicherbereich Reihen*Spalten
 } Matrix;*/
-Matrix createMatrix(unsigned int rows, unsigned int cols) {
-  Matrix matrix;
-  Matrix errorMatrix = {0, 0, NULL};
-  if (rows == 0 || cols == 0) {
 
+Matrix createMatrix(const unsigned int rows, const unsigned int cols) {
+  if (cols == 0 || rows == 0) {
+    Matrix errorMatrix = {0, 0, NULL};
     return errorMatrix;
   }
-  matrix.rows = rows;
-  matrix.cols = cols;
-
-  matrix.buffer = malloc(rows * cols * sizeof(MatrixType));
-  if (matrix.buffer == NULL) {
-    matrix.rows = 0;
-    matrix.cols = 0;
-    return matrix;
-  }
-
-  for (int i = 0; i < rows; i++) {
-    for (int j = 0; j < cols; j++) {
-      matrix.buffer[i * matrix.cols + j] = UNDEFINED_MATRIX_VALUE;
-    }
-  }
-  return matrix;
+  MatrixType *buffer =
+      malloc(rows * cols * sizeof(MatrixType)); // Speicher reservieren, malloc
+                                                // liefert Zeiger auf Speicher
+  Matrix newMatrix = {rows, cols, buffer};      // neue Matrix nach struct
+  return newMatrix;
 }
 void clearMatrix(Matrix *matrix) {
 
@@ -47,18 +34,18 @@ void setMatrixAt(const MatrixType value, Matrix matrix,
                  const unsigned int rowIdx, // Kopie der Matrix wird übergeben
                  const unsigned int colIdx) {
 
-  if (rowIdx >= matrix.rows || colIdx >= matrix.cols ||
-      matrix.buffer == NULL) { // Speichergröße nicht überschreiten
+  if (rowIdx >= matrix.rows || colIdx >= matrix.cols) {
+    // Speichergröße nicht überschreiten
     return;
   }
-
-  matrix.buffer[rowIdx * matrix.cols + colIdx] =
-      value; // rowIdx * matrix.cols -> Beginn der Zeile colIdx ->Spalte
-             // innerhalb der Zeile
+  matrix.buffer[rowIdx * matrix.cols + colIdx] = value;
+  // rowIdx * matrix.cols -> Beginn der Zeile colIdx ->Spalte
+  // innerhalb der Zeile
 }
-MatrixType getMatrixAt(const Matrix matrix,
-                       unsigned int rowIdx, // Kopie der Matrix wird übergeben
-                       unsigned int colIdx) {
+MatrixType
+getMatrixAt(const Matrix matrix,
+            const unsigned int rowIdx, // Kopie der Matrix wird übergeben
+            const unsigned int colIdx) {
   if (rowIdx >= matrix.rows || colIdx >= matrix.cols ||
       matrix.buffer == NULL) { // Speichergröße nicht überschreiten
     return UNDEFINED_MATRIX_VALUE;
@@ -68,187 +55,134 @@ MatrixType getMatrixAt(const Matrix matrix,
 
   return value;
 }
-
-Matrix broadCastCols(const Matrix matrix, const unsigned int rows,
-                     const unsigned int cols) {
-
-  Matrix copy = createMatrix(
-      rows, cols); // Matrix 1 Kopie erstellen mit Dimensionen von Matrix2
-  for (int r = 0; r < rows; r++) {
-
-    MatrixType value = getMatrixAt(matrix, r, 0);
+Matrix broadCastCols(const Matrix matrix, const unsigned int cols) {
+  Matrix copy1 = createMatrix(matrix.rows, cols);
+  for (int r = 0; r < matrix.rows; r++) {
+    MatrixType valueMatrix1 = getMatrixAt(matrix, r, 0);
     for (int c = 0; c < cols; c++) {
-
-      setMatrixAt(value, copy, r, c);
+      setMatrixAt(valueMatrix1, copy1, r, c);
     }
   }
-
-  return copy;
+  return copy1;
 }
-Matrix broadCastRows(const Matrix matrix, const unsigned int rows,
-                     const unsigned int cols) {
-
-  Matrix copy = createMatrix(rows, cols);
-
-  for (int c = 0; c < cols; c++) {
-    MatrixType value = getMatrixAt(matrix, 0, c);
-
+Matrix broadCastRows(const Matrix matrix, const unsigned int rows) {
+  Matrix copy1 = createMatrix(rows, matrix.cols);
+  for (int c = 0; c < matrix.cols; c++) {
+    MatrixType valueMatrix1 = getMatrixAt(matrix, 0, c);
     for (int r = 0; r < rows; r++) {
-
-      setMatrixAt(value, copy, r, c);
+      setMatrixAt(valueMatrix1, copy1, r, c);
     }
   }
-
-  return copy;
+  return copy1;
 }
-
 Matrix add(const Matrix matrix1, const Matrix matrix2) {
 
-  const unsigned int rows1 = matrix1.rows;
-  const unsigned int rows2 = matrix2.rows;
-  const unsigned int cols1 = matrix1.cols;
-  const unsigned int cols2 = matrix2.cols;
+  // Ergebnismatrix
+  Matrix result;
+  const int cols1 = matrix1.cols;
+  const int rows1 = matrix1.rows;
+  const int cols2 = matrix2.cols;
+  const int rows2 = matrix2.rows;
 
-  const int rowsEqual = ((rows1 == rows2) ? 1 : 0);
+  const int rowsEqual = (matrix1.rows == matrix2.rows) ? 1 : 0;
+  const int colsEqual = (matrix1.cols == matrix2.cols) ? 1 : 0;
 
-  const int colsEqual = ((cols1 == cols2) ? 1 : 0);
-
-  if (rowsEqual && colsEqual) // addieren
-
-  {
-    Matrix result = createMatrix(rows1, cols1); // Speicher reservieren
+  // Broadcasting nur bei Vektor und Matrix, Fehlermeldung bei zwei unpassender
+  // Matrix
+  if (rowsEqual == 1 && colsEqual == 1) {
+    Matrix result = createMatrix(matrix1.rows, matrix1.cols);
     if (result.buffer == NULL) {
       return (Matrix){0, 0, NULL};
     }
-
-    for (int i = 0; i < (rows1 * cols1); i++) { // addieren
-
-      result.buffer[i] =
-          (matrix1.buffer[i] +
-           matrix2.buffer[i]); // buffer[i]  ⇔  *(buffer + i) Adresse =
-                               // Startadresse + (i * sizeof(MatrixType))
+    for (int i = 0; i < rows1; i++) {
+      for (int j = 0; j < cols1; j++) {
+        int valueM1 = getMatrixAt(matrix1, i, j);
+        int valueM2 = getMatrixAt(matrix2, i, j);
+        int sum = valueM1 + valueM2;
+        setMatrixAt(sum, result, i, j);
+      }
     }
-
-    return result; // zurückgeben
-  }
-
-  else if (rowsEqual && !colsEqual) {
-
-    if (cols1 == 1) {
-
-      Matrix result = createMatrix(rows2, cols2);
+    return result;
+  } else if (rowsEqual == 1 && (cols1 == 1 || cols2 == 1)) {
+    if (cols1 == 1) { // broadcasting von vektor 1 zu matrix 1, add
+      Matrix newMatrix = broadCastCols(matrix1, cols2);
+      // add
+      Matrix result = createMatrix(newMatrix.rows, newMatrix.cols);
       if (result.buffer == NULL) {
         return (Matrix){0, 0, NULL};
       }
-
-      Matrix copy1 = broadCastCols(matrix1, rows2, cols2);
-      if (!copy1.buffer) {
-        clearMatrix(&result);
+      for (int i = 0; i < rows1; i++) {
+        for (int j = 0; j < cols2; j++) {
+          int valueM1 = getMatrixAt(newMatrix, i, j);
+          int valueM2 = getMatrixAt(matrix2, i, j);
+          int sum = valueM1 + valueM2;
+          setMatrixAt(sum, result, i, j);
+        }
+      }
+      clearMatrix(&newMatrix);
+      return result;
+    } else {
+      Matrix newMatrix2 = broadCastCols(matrix2, cols1);
+      // add
+      Matrix result = createMatrix(newMatrix2.rows, newMatrix2.cols);
+      if (result.buffer == NULL) {
         return (Matrix){0, 0, NULL};
       }
-
-      for (unsigned int i = 0; i < rows2 * cols2; i++) {
-        result.buffer[i] = copy1.buffer[i] + matrix2.buffer[i];
+      for (int i = 0; i < rows1; i++) {
+        for (int j = 0; j < cols1; j++) {
+          int valueM1 = getMatrixAt(matrix1, i, j);
+          int valueM2 = getMatrixAt(newMatrix2, i, j);
+          int sum = valueM1 + valueM2;
+          setMatrixAt(sum, result, i, j);
+        }
       }
 
-      /* freigeben, weil nicht mehr benötigt */
-      clearMatrix(&copy1);
-      return result;
-
-      // add und return
-
-    } else if (cols2 == 1) {
-
-      Matrix result = createMatrix(rows1, cols1);
-      if (result.buffer == NULL) {
-        Matrix error = {0, 0, NULL};
-        return error;
-      }
-
-      // Matrix2 hat nur eine Spalte -> horizontal broadcasten
-      Matrix copy2 = broadCastCols(matrix2, rows1, cols1);
-
-      for (unsigned int i = 0; i < rows1 * cols1; i++) {
-        result.buffer[i] = matrix1.buffer[i] + copy2.buffer[i];
-      }
-
-      // Optional: Speicher von copy2 freigeben
-      clearMatrix(&copy2);
-
       return result;
     }
-
-    else {
-
-      printf("Fehlermeldung"); // vielleicht Fehlermeldung ändern zu
-                               // Programmabbruch
-      Matrix error = {0, 0, NULL};
-      return error;
-    }
-
   }
 
-  else if (!rowsEqual && colsEqual) {
-
+  else if ((rows1 == 1 || rows2 == 1) && colsEqual == 1) {
     if (rows1 == 1) {
-
-      Matrix result = createMatrix(rows2, cols2);
+      Matrix newMatrix = broadCastRows(matrix1, rows2);
+      // add
+      Matrix result = createMatrix(newMatrix.rows, newMatrix.cols);
       if (result.buffer == NULL) {
         return (Matrix){0, 0, NULL};
       }
-
-      Matrix copy1 = broadCastRows(matrix1, rows2, cols2);
-
-      for (int i = 0; i < (rows2 * cols2); i++) { // addieren
-
-        result.buffer[i] =
-            (copy1.buffer[i] +
-             matrix2.buffer[i]); // buffer[i]  ⇔  *(buffer + i) Adresse =
-                                 // Startadresse + (i * sizeof(MatrixType))
+      for (int i = 0; i < rows2; i++) {
+        for (int j = 0; j < cols1; j++) {
+          int valueM1 = getMatrixAt(newMatrix, i, j);
+          int valueM2 = getMatrixAt(matrix2, i, j);
+          int sum = valueM1 + valueM2;
+          setMatrixAt(sum, result, i, j);
+        }
       }
       return result;
-
-      // add und return
-
-    } else if (rows2 == 1) {
-
-      Matrix result = createMatrix(rows1, cols1);
+    } else {
+      Matrix newMatrix2 = broadCastRows(matrix2, rows1);
+      // add
+      Matrix result = createMatrix(newMatrix2.rows, newMatrix2.cols);
       if (result.buffer == NULL) {
         return (Matrix){0, 0, NULL};
       }
-
-      Matrix copy2 = broadCastRows(matrix2, rows1, cols1);
-
-      // add und return
-
-      for (int i = 0; i < (rows1 * cols1); i++) { // addieren
-
-        result.buffer[i] =
-            (matrix1.buffer[i] +
-             copy2.buffer[i]); // buffer[i]  ⇔  *(buffer + i) Adresse =
-                               // Startadresse + (i * sizeof(MatrixType))
+      for (int i = 0; i < rows1; i++) {
+        for (int j = 0; j < cols1; j++) {
+          int valueM1 = getMatrixAt(matrix1, i, j);
+          int valueM2 = getMatrixAt(newMatrix2, i, j);
+          int sum = valueM1 + valueM2;
+          setMatrixAt(sum, result, i, j);
+        }
       }
+      clearMatrix(&newMatrix2);
       return result;
-
-    }
-
-    else {
-
-      printf("Fehlermeldung"); // vielleicht Fehlermeldung ändern zu
-                               // Programmabbruch
-      Matrix error = {0, 0, NULL};
-      return error;
     }
+  } else {
+    // kein add möglich
+    Matrix errorMatrix = {0, 0, NULL};
+    return errorMatrix;
   }
-
-  else {
-    printf("Fehlermeldung"); // vielleicht Fehlermeldung ändern zu
-                             // Programmabbruch
-    Matrix error = {0, 0, NULL};
-    return error;
-  }
+  return result;
 }
-
 Matrix multiply(const Matrix matrix1, const Matrix matrix2) {
   // Spalten1 müssen gleich zeilen2 sein! dann multiplizieren
   if (matrix1.cols == matrix2.rows) {

matrix.h

@@ -13,17 +13,15 @@ typedef struct {
 
 } Matrix;
 
-Matrix createMatrix(unsigned int rows, unsigned int cols);
+Matrix createMatrix(const unsigned int rows, const unsigned int cols);
 void clearMatrix(Matrix *matrix);
-void setMatrixAt(MatrixType value, Matrix matrix, unsigned int rowIdx,
-                 unsigned int colIdx);
-MatrixType getMatrixAt(const Matrix matrix, unsigned int rowIdx,
-                       unsigned int colIdx);
+void setMatrixAt(const MatrixType value, Matrix matrix,
+                 const unsigned int rowIdx, const unsigned int colIdx);
+MatrixType getMatrixAt(const Matrix matrix, const unsigned int rowIdx,
+                       const unsigned int colIdx);
 
-Matrix broadCastCols(const Matrix matrix, const unsigned int rows,
-                     const unsigned int cols);
-Matrix broadCastRows(const Matrix matrix, const unsigned int rows,
-                     const unsigned int cols);
+Matrix broadCastCols(const Matrix matrix, const unsigned int cols);
+Matrix broadCastRows(const Matrix matrix, const unsigned int rows);
 Matrix add(const Matrix matrix1, const Matrix matrix2);
 Matrix multiply(const Matrix matrix1, const Matrix matrix2);
 

neuralNetworkTests.c

@@ -28,68 +28,72 @@ Gewichte: bestimmen, wie stark ein Eingangssignal auf ein Neuron wirkt
 
 Dimension: Form der Matrizen für einen Layer*/
 
-// speichert NeuralNetwork nn in binäre Datei->erzeugt Dateiformat
+/* Gewichtsmatrix der Layer:
+ */
+
+// speichert NeuralNetwork nn in binäre Datei->später kann es wieder geöffnet
+// werden
 static void prepareNeuralNetworkFile(const char *path, const NeuralNetwork nn) {
-  FILE *f = fopen(path, "wb"); // Binärdatei zum Schreiben öffnen
-  if (f == NULL)
-    return;
+  FILE *fptr = fopen(path, "wb"); // Binärdatei zum Schreiben öffnen
+  if (fptr == NULL)
+    return; // file konnte nicht geöffnet werden
 
   // Header ist Erkennungsstring am Anfang der Datei, loadmodel erkennt
   // Dateiformat
-  const char header[] = "__info2_neural_network_file_format__";
-  fwrite(header, sizeof(char), strlen(header), f);
+  const char header[] = "__info2_neural_network_file_format__"; // header string
+  fwrite(header, sizeof(char), strlen(header),
+         fptr); // der header wird am Anfang der Datei platziert
 
-  // Wenn es keine Layer gibt, 0 eintragen, LoadModel gibt 0 zurück
+  // Wenn es keine Layer gibt, 0 eintragen, LoadModel erkennt, dass Datei leer
+  // ist
   if (nn.numberOfLayers == 0) {
     int zero = 0;
-    fwrite(&zero, sizeof(int), 1, f);
-    fclose(f);
+    fwrite(&zero, sizeof(int), 1, fptr);
+    fclose(fptr);
     return;
   }
 
   // Layer 0, inputDimension: Anzahl Input-Neuronen, outputDimension: Anzahl
-  // Output-Neuronen
+  // Output-Neuronen wird in Datei eingefügt
   int inputDim = (int)nn.layers[0].weights.cols;
   int outputDim = (int)nn.layers[0].weights.rows;
-  fwrite(&inputDim, sizeof(int), 1, f);
-  fwrite(&outputDim, sizeof(int), 1, f);
+  fwrite(&inputDim, sizeof(int), 1, fptr);
+  fwrite(&outputDim, sizeof(int), 1, fptr);
 
   /* 3) Für jede Layer in Reihenfolge: Gewichte (output x input), Biases (output
      x 1). Zwischen Layern wird nur die nächste outputDimension (int)
      geschrieben. */
   for (int i = 0; i < nn.numberOfLayers; i++) {
-    Layer layer = nn.layers[i];
+    Layer layer = nn.layers[i]; // kürzer, durch alle layer iterieren
 
     int wrows = (int)layer.weights.rows;
     int wcols = (int)layer.weights.cols;
-    int wcount = wrows * wcols;
+    int wcount = wrows * wcols; // Anzahl Gewichtseinträge
     int bcount =
-        layer.biases.rows * layer.biases.cols; /* normalerweise rows * 1 */
+        layer.biases.rows * layer.biases.cols; // Anzahl der Bias-Einträge
 
-    /* Gewichte (MatrixType binär) */
+    /* Gewichte */
     if (wcount > 0 && layer.weights.buffer != NULL) {
-      fwrite(layer.weights.buffer, sizeof(MatrixType), (size_t)wcount, f);
-    }
+      fwrite(layer.weights.buffer, sizeof(MatrixType), (size_t)wcount, fptr);
+    } // Gewichte werden als Matrix gespeichert
 
-    /* Biases (MatrixType binär) */
+    /* Biases */
     if (bcount > 0 && layer.biases.buffer != NULL) {
-      fwrite(layer.biases.buffer, sizeof(MatrixType), (size_t)bcount, f);
-    }
+      fwrite(layer.biases.buffer, sizeof(MatrixType), (size_t)bcount, fptr);
+    } // Biases werden als Vektor gespeichert
 
-    /* Für die nächste Layer: falls vorhanden, schreibe deren outputDimension */
+    /* outputDimensionen der nächsten Layer */
     if (i + 1 < nn.numberOfLayers) {
       int nextOutput = (int)nn.layers[i + 1].weights.rows;
-      fwrite(&nextOutput, sizeof(int), 1, f);
+      fwrite(&nextOutput, sizeof(int), 1, fptr);
     } else {
-      /* Letzte Layer: wir können das Ende signalisieren, indem wir ein 0
-         schreiben. loadModel liest dann outputDimension = 0 und beendet die
-         Schleife. */
+      // loadModel erkennt 0 als Ende der Datei
       int zero = 0;
-      fwrite(&zero, sizeof(int), 1, f);
+      fwrite(&zero, sizeof(int), 1, fptr);
     }
   }
 
-  fclose(f);
+  fclose(fptr); // Datei schließen
 }
 
 void test_loadModelReturnsCorrectNumberOfLayers(void) {