From 8ce167ff0dddf59d5ef3e85f18f08a36fcc8d4ad Mon Sep 17 00:00:00 2001 From: gurkenhabicht Date: Thu, 24 May 2018 11:02:00 +0200 Subject: [PATCH] graphing is opt in now. --- src/ansi_c_implementation/NLMSvariants.c | 18 +- src/ansi_c_implementation/README.md | 10 +- src/cpp_implementation/NLMSvariants.cpp | 487 ++++++++++++----------- src/cpp_implementation/README.md | 10 +- 4 files changed, 268 insertions(+), 257 deletions(-) diff --git a/src/ansi_c_implementation/NLMSvariants.c b/src/ansi_c_implementation/NLMSvariants.c index 6175c74..609611d 100644 --- a/src/ansi_c_implementation/NLMSvariants.c +++ b/src/ansi_c_implementation/NLMSvariants.c @@ -64,13 +64,14 @@ int main( int argc, char **argv ) { char *colorChannel = (char *) malloc(sizeof(char)* 32); char *inputfile = (char *)malloc(sizeof(char) * 32); unsigned *seed = NULL; - unsigned k, xclude = 0; + unsigned k, include = 0; unsigned windowSize = 5; unsigned samplesCount = 512; char *stdcolor = "green", xBuffer[512]; colorChannel = stdcolor; unsigned int uint_buffer[1], windowBuffer[1]; double learnrate = 0.4; + char *istrue = "true"; while( (argc > 1) && (argv[1][0] == '-') ) { // Parses parameters from stdin @@ -111,9 +112,14 @@ int main( int argc, char **argv ) { ++argv; --argc; break; - case 'x': + case 'g': sscanf(&argv[1][3], "%s", xBuffer); - xclude = 1; + if ( strstr(xBuffer, istrue) ) { + include = 1; + } else { + printf( "Wrong Argruments: %s\n", argv[1]); + usage(argv); + } ++argv; --argc; break; @@ -169,7 +175,7 @@ int main( int argc, char **argv ) { directPredecessor ( mlData, points); differentialPredecessor( mlData, points ); - if ( xclude == 0 ) { + if ( include == 1 ) { mkSvgGraph(points); // Graph building } @@ -480,8 +486,8 @@ char * fileSuffix ( int id ) { "_localMean.txt", "_testvalues.txt", "_differential_predecessor.txt", - "_weights_used_local_mean", - "_weights_used_diff_pred", + "_weights_used_local_mean.txt", + "_weights_used_diff_pred.txt", }; return suffix[id]; } diff --git a/src/ansi_c_implementation/README.md b/src/ansi_c_implementation/README.md index bccb266..e667537 100644 --- a/src/ansi_c_implementation/README.md +++ b/src/ansi_c_implementation/README.md @@ -24,12 +24,12 @@ There are a bunch of options you can predefine but do not have to. The only para | Parameter | Description | StdVal | |:----------|:-----------------------------:|:-------| -| -i | The inputfile, has to be PPM | none | -| -n | Amount of input data used | 500 | -| -w | Size of M (window) | 5 | +| -i | The inputfile, has to be PPM. | none | +| -n | Amount of input data used. | 500 | +| -w | Size of M (window). | 5 | | -c | Choose RGB color channel, green has least noise. | green | -| -l | Learnrate of machine learning | 0.4 | -| -x | Exclude graph building. Logfiles only, choose for insane amount of input data. 10Mio. Pixels tested so far.| none| +| -l | Learnrate of machine learning.| 0.4 | +| -g true | include graph building. Choose for amount of input data lower than 1200.| none| | -s | Seed randomizing weights. Choose for repoducability. | time(NULL)| This code is ANSI compatible no POSIX, C99, C11 or GNU libs, because it had to be VS compatible . There are way easier methods like getline() or getopt(), I know ... diff --git a/src/cpp_implementation/NLMSvariants.cpp b/src/cpp_implementation/NLMSvariants.cpp index 806ec2e..609611d 100644 --- a/src/cpp_implementation/NLMSvariants.cpp +++ b/src/cpp_implementation/NLMSvariants.cpp @@ -23,116 +23,122 @@ double *xSamples; // Input values mldata_t *mlData = NULL; // Machine learning point_t *points = NULL; // Graphing - /* *graph building* */ +/* *graph building* */ static imagePixel_t * rdPPM(char *fileName); // Read PPM file format void mkPpmFile(char *fileName, imagePixel_t *image); // Writes PPM file int ppmColorChannel(FILE* fp, imagePixel_t *image, // Writes colorChannel from PPM file to log file - char *colorChannel, mldata_t *mlData); + char *colorChannel, mldata_t *mlData); void colorSamples(FILE* fp, mldata_t *mlData); // Stores color channel values in xSamples - /* *file handling* */ -char * mkFileName(char* buffer, - size_t max_len, int suffixId); -char *fileSuffix(int id); -char *fileHeader(int id); // Header inside the logfiles - //void myLogger ( FILE* fp, point_t points[] ); +/* *file handling* */ +char * mkFileName ( char* buffer, + size_t max_len, int suffixId ); +char *fileSuffix ( int id ); +char *fileHeader ( int id ); // Header inside the logfiles +//void myLogger ( FILE* fp, point_t points[] ); void bufferLogger(char *buffer, point_t points[]); // Writes points to graph template -void mkSvgGraph(point_t points[]); // Parses graph template and calls bufferLogger() -void weightsLogger(double *weights, int suffix); // Writes updated weights to a file +void mkSvgGraph ( point_t points[] ); // Parses graph template and calls bufferLogger() +void weightsLogger ( double *weights, int suffix ); // Writes updated weights to a file - /* *rand seed* */ -double r2(void); // Random val between 0 and 1 -double rndm(void); +/* *rand seed* */ +double r2 ( void ); // Random val between 0 and 1 +double rndm ( void ); /* *args parser* */ -void usage(char **argv); // Help text called by args parser +void usage ( char **argv ); // Help text called by args parser - /* *math* */ +/* *math* */ mldata_t * init_mldata_t(unsigned windowSize, unsigned samplesCount, double learnrate); double sum_array(double x[], int length); -void localMean(mldata_t *mlData, point_t points[]); // First, -void directPredecessor(mldata_t *mlData, point_t points[]); // Second, -void differentialPredecessor(mldata_t *mlData, point_t points[]); // Third filter implementation +void localMean ( mldata_t *mlData,point_t points[] ); // First, +void directPredecessor ( mldata_t *mlData, point_t points[] ); // Second, +void differentialPredecessor ( mldata_t *mlData, point_t points[] ); // Third filter implementation double *popNAN(double *xError); // Returns array without NAN values, if any exist double windowXMean(int _arraylength, int xCount); // Returns mean value of given window -int main(int argc, char **argv) { - char *colorChannel = (char *)malloc(sizeof(char) * 32); +int main( int argc, char **argv ) { + char *colorChannel = (char *) malloc(sizeof(char)* 32); char *inputfile = (char *)malloc(sizeof(char) * 32); unsigned *seed = NULL; - unsigned k, xclude = 0; + unsigned k, include = 0; unsigned windowSize = 5; unsigned samplesCount = 512; char *stdcolor = "green", xBuffer[512]; colorChannel = stdcolor; unsigned int uint_buffer[1], windowBuffer[1]; double learnrate = 0.4; - - - while ((argc > 1) && (argv[1][0] == '-')) { // Parses parameters from stdin - switch (argv[1][1]) { - case 'i': - inputfile = &argv[1][3]; - ++argv; - --argc; - break; - case 'w': - sscanf(&argv[1][3], "%u", windowBuffer); - windowSize = windowBuffer[0]; - ++argv; - --argc; - break; - case 'c': - colorChannel = &argv[1][3]; - ++argv; - --argc; - break; - case 's': - sscanf(&argv[1][3], "%u", uint_buffer); - seed = &uint_buffer[0]; - ++argv; - --argc; - break; - case 'n': - sscanf(&argv[1][3], "%u", &samplesCount); - ++argv; - --argc; - break; - case'h': - printf("Program name: %s\n", argv[0]); - usage(argv); - break; - case 'l': - sscanf(&argv[1][3], "%lf", &learnrate); - ++argv; - --argc; - break; - case 'x': - sscanf(&argv[1][3], "%s", xBuffer); - xclude = 1; - ++argv; - --argc; - break; - default: - printf("Wrong Arguments: %s\n", argv[1]); - usage(argv); - } - + char *istrue = "true"; + + + while( (argc > 1) && (argv[1][0] == '-') ) { // Parses parameters from stdin + switch( argv[1][1] ) { + case 'i': + inputfile = &argv[1][3]; + ++argv; + --argc; + break; + case 'w': + sscanf(&argv[1][3], "%u", windowBuffer); + windowSize = windowBuffer[0]; + ++argv; + --argc; + break; + case 'c': + colorChannel = &argv[1][3]; + ++argv; + --argc; + break; + case 's': + sscanf(&argv[1][3], "%u", uint_buffer); + seed = &uint_buffer[0]; + ++argv; + --argc; + break; + case 'n': + sscanf(&argv[1][3], "%u", &samplesCount); + ++argv; + --argc; + break; + case'h': + printf("Program name: %s\n", argv[0]); + usage(argv); + break; + case 'l': + sscanf(&argv[1][3], "%lf", &learnrate); + ++argv; + --argc; + break; + case 'g': + sscanf(&argv[1][3], "%s", xBuffer); + if ( strstr(xBuffer, istrue) ) { + include = 1; + } else { + printf( "Wrong Argruments: %s\n", argv[1]); + usage(argv); + } + ++argv; + --argc; + break; + default: + printf("Wrong Arguments: %s\n", argv[1]); + usage(argv); + } + ++argv; --argc; } - init_mldata_t(windowSize, samplesCount, learnrate); - xSamples = (double *)malloc(sizeof(double) * mlData->samplesCount); // Resize input values - points = (point_t *)malloc(sizeof(point_t) * mlData->samplesCount); // Resize points - imagePixel_t *image; + init_mldata_t ( windowSize, samplesCount, learnrate ); + xSamples = (double *) malloc ( sizeof(double) * mlData->samplesCount ); // Resize input values + points = (point_t *) malloc ( sizeof(point_t) * mlData->samplesCount); // Resize points + imagePixel_t *image; image = rdPPM(inputfile); // Set Pointer on input values - printf("window Size: %d\n", mlData->windowSize); + printf("window Size: %d\n", mlData->windowSize); char fileName[50]; // Logfiles and their names mkFileName(fileName, sizeof(fileName), TEST_VALUES); FILE* fp5 = fopen(fileName, "w"); @@ -141,17 +147,16 @@ int main(int argc, char **argv) { FILE* fp6 = fopen(fileName, "r"); colorSamples(fp6, mlData); - if ((seed != NULL)) { - srand(*seed); // Seed for random number generating + if ( (seed != NULL) ){ + srand( *seed ); // Seed for random number generating printf("srand is reproducable\n"); - } - else { - srand((unsigned int)time(NULL)); + } else { + srand( (unsigned int)time(NULL) ); printf("srand depends on time\n"); // Default seed is time(NULL) } printf("generated weights:\n"); - for (k = 0; k < mlData->windowSize; k++) { + for (k = 0; k < mlData->windowSize; k++) { mlData->weights[k] = rndm(); // Init random weights printf("[%d] %lf\n", k, mlData->weights[k]); } @@ -160,20 +165,20 @@ int main(int argc, char **argv) { mkFileName(fileName, sizeof(fileName), PURE_WEIGHTS); // Logfile weights FILE *fp0 = fopen(fileName, "w"); for (k = 0; k < mlData->windowSize; k++) { - fprintf(fp0, "[%d]%lf\n", k, mlData->weights[k]); + fprintf(fp0, "[%d]%lf\n", k, mlData->weights[k]); } fclose(fp0); /* *math magic* */ - localMean(mlData, points); - directPredecessor(mlData, points); - differentialPredecessor(mlData, points); + localMean ( mlData, points ); + directPredecessor ( mlData, points); + differentialPredecessor( mlData, points ); - if (xclude == 0) { + if ( include == 1 ) { mkSvgGraph(points); // Graph building - - } + + } free(image); free(xSamples); @@ -191,56 +196,56 @@ Variant (1/3), substract local mean. ====================================================================================================== */ -void localMean(mldata_t *mlData, point_t points[]) { - double *localWeights = (double *)malloc(sizeof(double) * mlData->windowSize + 1); - localWeights = mlData->weights; // Copy weights so they can be changed locally +void localMean ( mldata_t *mlData, point_t points[] ) { + double *localWeights = (double *) malloc ( sizeof(double) * mlData->windowSize + 1); + localWeights = mlData->weights; char fileName[50]; - double *xError = (double *)malloc(sizeof(double) * mlData->samplesCount + 1); // Includes e(n) + double *xError = (double *) malloc ( sizeof(double) * mlData->samplesCount + 1); // Includes e(n) memset(xError, 0.0, mlData->samplesCount); // Initialize xError-array with Zero - unsigned i, xCount = 0; // Runtime vars + unsigned i, xCount = 0; // Runtime vars mkFileName(fileName, sizeof(fileName), LOCAL_MEAN); // Create Logfile and its filename - FILE* fp4 = fopen(fileName, "w"); - fprintf(fp4, fileHeader(LOCAL_MEAN_HEADER)); - - mkFileName(fileName, sizeof(fileName), USED_WEIGHTS_LOCAL_MEAN); + FILE* fp4 = fopen(fileName, "w"); + fprintf( fp4, fileHeader(LOCAL_MEAN_HEADER) ); + + mkFileName ( fileName, sizeof(fileName), USED_WEIGHTS_LOCAL_MEAN); FILE *fp9 = fopen(fileName, "w"); - + double xMean = xSamples[0]; double xSquared = 0.0; double xPredicted = 0.0; double xActual = 0.0; - for (xCount = 1; xCount < mlData->samplesCount-1; xCount++) { // First value will not get predicted - unsigned _arrayLength = (xCount > mlData->windowSize) ? mlData->windowSize + 1 : xCount; // Ensures corect length at start - xMean = (xCount > 0) ? windowXMean(_arrayLength, xCount) : 0; + for ( xCount = 1; xCount < mlData->samplesCount-1; xCount++ ) { // First value will not get predicted + unsigned _arrayLength = ( xCount > mlData->windowSize ) ? mlData->windowSize + 1 : xCount; // Ensures corect length at start + xMean = (xCount > 0) ? windowXMean(_arrayLength, xCount) : 0; xPredicted = 0.0; xActual = xSamples[xCount]; - for (i = 1; i < _arrayLength; i++) { // Get predicted value - xPredicted += (localWeights[i - 1] * (xSamples[xCount - i] - xMean)); + for ( i = 1; i < _arrayLength; i++ ) { // Get predicted value + xPredicted += ( localWeights[i - 1] * (xSamples[xCount - i] - xMean) ); } - xPredicted += xMean; + xPredicted += xMean; xError[xCount] = xActual - xPredicted; // Get error value xSquared = 0.0; for (i = 1; i < _arrayLength; i++) { // Get xSquared xSquared += pow(xSamples[xCount - i] - xMean, 2); } - if (xSquared == 0.0) { // Otherwise returns Pred: -1.#IND00 in some occassions + if ( xSquared == 0.0 ) { // Otherwise returns Pred: -1.#IND00 in some occassions xSquared = 1.0; } - for (i = 1; i < _arrayLength; i++) { // Update weights + for ( i = 1; i < _arrayLength; i++ ) { // Update weights localWeights[i] = localWeights[i - 1] + mlData->learnrate * xError[xCount] // Substract localMean - * ((xSamples[xCount - i] - xMean) / xSquared); - fprintf(fp9, "%lf\n", localWeights[i]); + * ( (xSamples[xCount - i] - xMean) / xSquared ); + fprintf( fp9, "%lf\n", localWeights[i] ); } - + fprintf(fp4, "%d\t%f\t%f\t%f\n", xCount, xPredicted, xActual, xError[xCount]); // Write to logfile - + points[xCount].xVal[1] = xCount; // Save points so graph can be build later on - points[xCount].yVal[1] = xPredicted; + points[xCount].yVal[1] = xPredicted; points[xCount].xVal[4] = xCount; points[xCount].yVal[4] = xError[xCount]; @@ -249,8 +254,8 @@ void localMean(mldata_t *mlData, point_t points[]) { fclose(fp9); double *xErrorPtr = popNAN(xError); // delete NAN values from xError[] double xErrorLength = *xErrorPtr; // Watch popNAN()! - xErrorPtr[0] = 0.0; - // printf("Xerrorl:%lf", xErrorLength); + xErrorPtr[0] = 0.0; +// printf("Xerrorl:%lf", xErrorLength); double mean = sum_array(xErrorPtr, xErrorLength) / xErrorLength; // Mean double deviation = 0.0; @@ -260,7 +265,7 @@ void localMean(mldata_t *mlData, point_t points[]) { deviation /= xErrorLength; // Deviation printf("mean:%lf, devitation:%lf\t\tlocal Mean\n", mean, deviation); fprintf(fp4, "\nQuadratische Varianz(x_error): %f\nMittelwert:(x_error): %f\n\n", deviation, mean); // Write to logfile - //free(localWeights); + free(localWeights); free(xErrorPtr); free(xError); @@ -279,11 +284,12 @@ substract direct predecessor ====================================================================================================== */ -void directPredecessor(mldata_t *mlData, point_t points[]) { - double *localWeights = (double *)malloc(sizeof(double) * mlData->windowSize + 1); +void directPredecessor( mldata_t *mlData, point_t points[]) { + double *localWeights = ( double * ) malloc ( sizeof(double) * mlData->windowSize + 1 ); localWeights = mlData->weights; + char fileName[512]; - double *xError = (double *)malloc(sizeof(double) * mlData->samplesCount + 1); + double *xError = (double *) malloc ( sizeof(double) * mlData->samplesCount + 1 ); memset(xError, 0.0, mlData->samplesCount); unsigned xCount = 0, i; double xActual = 0.0; @@ -291,20 +297,20 @@ void directPredecessor(mldata_t *mlData, point_t points[]) { mkFileName(fileName, sizeof(fileName), DIRECT_PREDECESSOR); // Logfile and name handling FILE *fp3 = fopen(fileName, "w"); - fprintf(fp3, fileHeader(DIRECT_PREDECESSOR_HEADER)); - - mkFileName(fileName, sizeof(fileName), USED_WEIGHTS_DIR_PRED); + fprintf( fp3, fileHeader(DIRECT_PREDECESSOR_HEADER) ); + + mkFileName ( fileName, sizeof(fileName), USED_WEIGHTS_DIR_PRED); FILE *fp9 = fopen(fileName, "w"); for (xCount = 1; xCount < mlData->samplesCount-1; xCount++) { // first value will not get predicted - unsigned _arrayLength = (xCount > mlData->windowSize) ? mlData->windowSize + 1 : xCount; + unsigned _arrayLength = ( xCount > mlData->windowSize ) ? mlData->windowSize + 1 : xCount; xPredicted = 0.0; xActual = xSamples[xCount]; - + for (i = 1; i < _arrayLength; i++) { - xPredicted += (localWeights[i - 1] * (xSamples[xCount - 1] - xSamples[xCount - i - 1])); + xPredicted += ( localWeights[i - 1] * (xSamples[xCount - 1] - xSamples[xCount - i - 1])); } - + xPredicted += xSamples[xCount - 1]; xError[xCount] = xActual - xPredicted; @@ -312,32 +318,32 @@ void directPredecessor(mldata_t *mlData, point_t points[]) { for (i = 1; i < _arrayLength; i++) { xSquared += pow(xSamples[xCount - 1] - xSamples[xCount - i - 1], 2); // substract direct predecessor } - if (xSquared == 0.0) { // Otherwise returns Pred: -1.#IND00 in some occassions + if ( xSquared == 0.0 ) { // Otherwise returns Pred: -1.#IND00 in some occassions xSquared = 1.0; } - for (i = 1; i < _arrayLength; i++) { // Update weights - localWeights[i] = localWeights[i - 1] + mlData->learnrate * xError[xCount] - * ((xSamples[xCount - 1] - xSamples[xCount - i - 1]) / xSquared); - fprintf(fp9, "%lf\n", localWeights[i]); + for ( i = 1; i < _arrayLength; i++ ) { // Update weights + localWeights[i] = localWeights[i-1] + mlData->learnrate * xError[xCount] + * ( (xSamples[xCount - 1] - xSamples[xCount - i - 1]) / xSquared); + fprintf( fp9, "%lf\n", localWeights[i] ); } - - fprintf(fp3, "%d\t%f\t%f\t%f\n", xCount, xPredicted, xActual, xError[xCount]); // Write to logfile + + fprintf(fp3, "%d\t%f\t%f\t%f\n", xCount, xPredicted, xActual, xError[xCount]); // Write to logfile points[xCount].xVal[2] = xCount; // Fill point_t array for graph building points[xCount].yVal[2] = xPredicted; points[xCount].xVal[5] = xCount; points[xCount].yVal[5] = xError[xCount]; - // weightsLogger( fp, localWeights, USED_WEIGHTS ); + // weightsLogger( fp, localWeights, USED_WEIGHTS ); } fclose(fp9); double *xErrorPtr = popNAN(xError); // delete NAN values from xError[] double xErrorLength = *xErrorPtr; // Watch popNAN()! - xErrorPtr[0] = 0.0; // Stored length in [0] , won't be used anyway. Bit dirty - //printf("Xerrorl:%lf", xErrorLength); + xErrorPtr[0] = 0.0; // Stored length in [0] , won't be used anyway. Bit dirty + //printf("Xerrorl:%lf", xErrorLength); double mean = sum_array(xErrorPtr, xErrorLength) / xErrorLength; // Mean - double deviation = 0.0; + double deviation = 0.0; for (i = 1; i < xErrorLength; i++) { @@ -347,7 +353,7 @@ void directPredecessor(mldata_t *mlData, point_t points[]) { printf("mean:%lf, devitation:%lf\t\tdirect Predecessor\n", mean, deviation); fprintf(fp3, "\nQuadratische Varianz(x_error): %f\nMittelwert:(x_error): %f\n\n", deviation, mean); fclose(fp3); - //free(localWeights); + free(localWeights); free(xErrorPtr); free(xError); } @@ -362,31 +368,33 @@ differential predecessor. ====================================================================================================== */ -void differentialPredecessor(mldata_t *mlData, point_t points[]) { - double *localWeights = (double *)malloc(sizeof(double) * mlData->windowSize + 1); +void differentialPredecessor ( mldata_t *mlData, point_t points[] ) { + double *localWeights = (double *) malloc ( sizeof(double) * mlData->windowSize + 1 ); localWeights = mlData->weights; + char fileName[512]; - double *xError = (double *)malloc(sizeof(double) * mlData->samplesCount + 1); + double *xError = (double *) malloc ( sizeof(double) * mlData->samplesCount + 1); memset(xError, 0.0, mlData->samplesCount); + unsigned xCount = 0, i; double xPredicted = 0.0; double xActual = 0.0; mkFileName(fileName, sizeof(fileName), DIFFERENTIAL_PREDECESSOR); // File handling FILE *fp6 = fopen(fileName, "w"); - fprintf(fp6, fileHeader(DIFFERENTIAL_PREDECESSOR_HEADER)); + fprintf(fp6, fileHeader(DIFFERENTIAL_PREDECESSOR_HEADER) ); - mkFileName(fileName, sizeof(fileName), USED_WEIGHTS_DIFF_PRED); + mkFileName ( fileName, sizeof(fileName), USED_WEIGHTS_DIFF_PRED); FILE *fp9 = fopen(fileName, "w"); - for (xCount = 1; xCount < mlData->samplesCount-1; xCount++) { // First value will not get predicted + for (xCount = 1; xCount < mlData->samplesCount-1; xCount++) { // First value will not get predicted unsigned _arrayLength = (xCount > mlData->windowSize) ? mlData->windowSize + 1 : xCount; xPredicted = 0.0; xActual = xSamples[xCount]; for (i = 1; i < _arrayLength; i++) { - xPredicted += (localWeights[i - 1] * (xSamples[xCount - i] - xSamples[xCount - i - 1])); + xPredicted += ( localWeights[i - 1] * (xSamples[xCount - i] - xSamples[xCount - i - 1])); } xPredicted += xSamples[xCount - 1]; xError[xCount] = xActual - xPredicted; @@ -395,18 +403,18 @@ void differentialPredecessor(mldata_t *mlData, point_t points[]) { for (i = 1; i < _arrayLength; i++) { xSquared += pow(xSamples[xCount - i] - xSamples[xCount - i - 1], 2); // Substract direct predecessor } - if (xSquared == 0.0) { // Otherwise returns Pred: -1.#IND00 in some occassions + if ( xSquared == 0.0 ) { // Otherwise returns Pred: -1.#IND00 in some occassions xSquared = 1.0; } for (i = 1; i < _arrayLength; i++) { - localWeights[i] = localWeights[i - 1] + mlData->learnrate * xError[xCount] + localWeights[i] = localWeights[i-1] + mlData->learnrate * xError[xCount] * ((xSamples[xCount - i] - xSamples[xCount - i - 1]) / xSquared); - fprintf(fp9, "%lf\n", localWeights[i]); + fprintf( fp9, "%lf\n", localWeights[i] ); } - fprintf(fp6, "%d\t%f\t%f\t%f\n", xCount, xPredicted, xActual, xError[xCount]); // Write to logfile - + fprintf(fp6, "%d\t%f\t%f\t%f\n", xCount, xPredicted, xActual, xError[xCount]); // Write to logfile + points[xCount].xVal[3] = xCount; points[xCount].yVal[3] = xPredicted; points[xCount].xVal[6] = xCount; @@ -416,13 +424,13 @@ void differentialPredecessor(mldata_t *mlData, point_t points[]) { fclose(fp9); double *xErrorPtr = popNAN(xError); // delete NAN values from xError[] double xErrorLength = *xErrorPtr; // Watch popNAN()! - xErrorPtr[0] = 0.0; - // printf("Xerrorl:%lf", xErrorLength); + xErrorPtr[0] = 0.0; +// printf("Xerrorl:%lf", xErrorLength); double mean = sum_array(xErrorPtr, xErrorLength) / xErrorLength; double deviation = 0.0; - + for (i = 1; i < xErrorLength; i++) { // Mean square deviation += pow(xError[i] - mean, 2); } @@ -430,12 +438,12 @@ void differentialPredecessor(mldata_t *mlData, point_t points[]) { printf("mean:%lf, devitation:%lf\t\tdifferential Predecessor\n", mean, deviation); fprintf(fp6, "\nQuadratische Varianz(x_error): %f\nMittelwert:(x_error): %f\n\n", deviation, mean); fclose(fp6); - //free(localWeights); + free(localWeights); free(xErrorPtr); free(xError); - // weightsLogger( localWeights, USED_WEIGHTS ); +// weightsLogger( localWeights, USED_WEIGHTS ); } /* @@ -453,7 +461,7 @@ char *mkFileName(char* buffer, size_t max_len, int suffixId) { const char * format_str = "%Y-%m-%d_%H_%M_%S"; // Date formatting size_t date_len; const char * suffix = fileSuffix(suffixId); - time_t now = time(NULL); + time_t now = time(NULL); strftime(buffer, max_len, format_str, localtime(&now)); // Get Date date_len = strlen(buffer); @@ -470,16 +478,16 @@ Contains and returns every suffix for all existing filenames ====================================================================================================== */ -char * fileSuffix(int id) { - char * suffix[] = { "_weights_pure.txt", - "_weights_used_dir_pred_.txt", - "_direct_predecessor.txt", - "_ergebnisse.txt", - "_localMean.txt", - "_testvalues.txt", - "_differential_predecessor.txt", - "_weights_used_local_mean", - "_weights_used_diff_pred", +char * fileSuffix ( int id ) { + char * suffix[] = { "_weights_pure.txt", + "_weights_used_dir_pred_.txt", + "_direct_predecessor.txt", + "_ergebnisse.txt", + "_localMean.txt", + "_testvalues.txt", + "_differential_predecessor.txt", + "_weights_used_local_mean.txt", + "_weights_used_diff_pred.txt", }; return suffix[id]; } @@ -489,14 +497,14 @@ char * fileSuffix(int id) { fileHeader -Contains and returns header from logfiles +Contains and returns header from logfiles ====================================================================================================== */ -char * fileHeader(int id) { - char * header[] = { "\n=========================== Local Mean ===========================\nNo.\txPredicted\txAcutal\t\txError\n", - "\n=========================== Direct Predecessor ===========================\nNo.\txPredicted\txAcutal\t\txError\n", - "\n=========================== Differential Predecessor ===========================\nNo.\txPredicted\txAcutal\t\txError\n" +char * fileHeader ( int id ) { + char * header[] = { "\n=========================== Local Mean ===========================\nNo.\txPredicted\txAcutal\t\txError\n", + "\n=========================== Direct Predecessor ===========================\nNo.\txPredicted\txAcutal\t\txError\n", + "\n=========================== Differential Predecessor ===========================\nNo.\txPredicted\txAcutal\t\txError\n" }; return header[id]; } @@ -510,7 +518,7 @@ Logs used weights to logfile ====================================================================================================== */ -void weightsLogger(double *weights, int val) { +void weightsLogger (double *weights, int val ) { char fileName[512]; unsigned i; mkFileName(fileName, sizeof(fileName), val); @@ -526,21 +534,21 @@ void weightsLogger(double *weights, int val) { bufferLogger -formats output of mkSvgGraph -- Please open graphResults.html to see the output-- -[0] = xActual, -[1] = xPredicted from localMean, -[2] = xPredicted from directPredecessor, -[3] = xPredicted from differentialpredecessor, -[4] = xError from localMean, -[5] = xError from directPredecessor, -[6] = xError from differentialPredecessor +formats output of mkSvgGraph -- Please open graphResults.html to see the output-- + [0] = xActual, + [1] = xPredicted from localMean, + [2] = xPredicted from directPredecessor, + [3] = xPredicted from differentialpredecessor, + [4] = xError from localMean, + [5] = xError from directPredecessor, + [6] = xError from differentialPredecessor ====================================================================================================== */ void bufferLogger(char *buffer, point_t points[]) { unsigned i; char _buffer[512] = ""; // TODO: resize buffer and _buffer so greater sampleval can be choosen - // char *_buffer = (char *) malloc ( sizeof(char) * 512 + 1); +// char *_buffer = (char *) malloc ( sizeof(char) * 512 + 1); for (i = 1; i < mlData->samplesCount - 1; i++) { // xActual sprintf(_buffer, "L %f %f\n", points[i].xVal[0], points[i].yVal[0]); strcat(buffer, _buffer); @@ -589,7 +597,7 @@ double sum_array(double x[], int xlength) { popNan -returns new array without NAN values +returns new array without NAN values ====================================================================================================== */ @@ -599,21 +607,21 @@ double *popNAN(double *xError) { double *tmp = NULL; double *more_tmp = NULL; - for (i = 0; i < mlData->samplesCount - 1; i++) { - counter++; - more_tmp = (double *)realloc(tmp, counter*(sizeof(double))); - if (!isnan(xError[i])) { - tmp = more_tmp; - tmp[counter - 1] = xError[i]; - //printf("xERROR:%lf\n", tmp[counter - 1]); - tmpLength++; - } + for ( i = 0; i < mlData->samplesCount - 1; i++ ) { + counter ++; + more_tmp = (double *) realloc ( tmp, counter*(sizeof(double) )); + if ( !isnan(xError[i]) ) { + tmp = more_tmp; + tmp[counter - 1] = xError[i]; + //printf("xERROR:%lf\n", tmp[counter - 1]); + tmpLength++; + } } counter += 1; - more_tmp = (double *)realloc(tmp, counter * sizeof(double)); + more_tmp = (double *) realloc ( tmp, counter * sizeof(double) ); tmp = more_tmp; *tmp = tmpLength; // Length of array is stored inside tmp[0]. tmp[0] is never used anyways - + return tmp; } @@ -668,14 +676,14 @@ void mkSvgGraph(point_t points[]) { fseek(input, 0, SEEK_END); long fpLength = ftell(input); fseek(input, 0, SEEK_SET); - + char buffer[131072] = ""; // Bit dirty - // char *buffer = (char *) malloc ( sizeof(char) * ( ( 3 * mlData->samplesCount ) + fpLength + 1 ) ); +// char *buffer = (char *) malloc ( sizeof(char) * ( ( 3 * mlData->samplesCount ) + fpLength + 1 ) ); memset(buffer, '\0', sizeof(buffer)); while (!feof(input)) { // parses file until "firstGraph" has been found - fgets(line, 512, input); + fgets(line, 512, input); strncat(buffer, line, strlen(line)); if (strstr(line, firstGraph) != NULL) { // Compares line <-> "firstGraph" bufferLogger(buffer, points); // write points @@ -723,30 +731,30 @@ static imagePixel_t *rdPPM(char *fileName) { c = getc(fp); } ungetc(c, fp); - if (fscanf(fp, "%d %d", &image->x, &image->y) != 2) { + if ( fscanf(fp, "%d %d", &image->x, &image->y) != 2 ) { fprintf(stderr, "Invalid image size in %s\n", fileName); exit(EXIT_FAILURE); } - if (fscanf(fp, "%d", &rgbColor) != 1) { + if ( fscanf(fp, "%d", &rgbColor) != 1 ) { fprintf(stderr, "Invalid rgb component in %s\n", fileName); } - if (rgbColor != RGB_COLOR) { + if ( rgbColor != RGB_COLOR ) { fprintf(stderr, "Invalid image color range in %s\n", fileName); exit(EXIT_FAILURE); } - while (fgetc(fp) != '\n'); + while ( fgetc(fp) != '\n' ); image->data = (colorChannel_t *)malloc(image->x * image->y * sizeof(imagePixel_t)); if (!image) { fprintf(stderr, "malloc() on image->data failed"); exit(EXIT_FAILURE); } - if ((image->x * image->y) < mlData->samplesCount) { - printf("Changing \"-n\" to %d, image max data size\n", (image->x * image->y)); - tmp = (double *)realloc(xSamples, sizeof(double) * (image->x * image->y)); + if ( (image->x * image->y) < mlData->samplesCount) { + printf("Changing \"-n\" to %d, image max data size\n", ( image->x * image->y ) ); + tmp = (double *) realloc ( xSamples, sizeof(double) * (image->x * image->y) ); xSamples = tmp; - mlData->samplesCount = (image->x * image->y) / sizeof(double); + mlData->samplesCount = (image->x * image->y ) / sizeof(double); } - if (fread(image->data, 3 * image->x, image->y, fp) != image->y) { + if ( fread( image->data, 3 * image->x, image->y, fp) != image->y) { fprintf(stderr, "Loading image failed"); exit(EXIT_FAILURE); } @@ -790,26 +798,23 @@ int ppmColorChannel(FILE* fp, imagePixel_t *image, char *colorChannel, mldata_t unsigned i = 0; printf("colorChannel : %s\n", colorChannel); - if (image) { // RGB channel can be set through args from cli - if (strcmp(colorChannel, "green") == 0) { - for (i = 0; i < mlData->samplesCount - 1; i++) { - fprintf(fp, "%d\n", image->data[i].green); + if ( image ) { // RGB channel can be set through args from cli + if ( strcmp(colorChannel, "green") == 0 ){ + for ( i = 0; i < mlData->samplesCount - 1; i++ ) { + fprintf ( fp, "%d\n", image->data[i].green ); } - } - else if (strcmp(colorChannel, "red") == 0) { - for (i = 0; i < mlData->samplesCount - 1; i++) { - fprintf(fp, "%d\n", image->data[i].red); + } else if ( strcmp(colorChannel, "red") == 0 ){ + for ( i = 0; i < mlData->samplesCount - 1; i++ ) { + fprintf ( fp, "%d\n", image->data[i].red ); + } + + } else if ( strcmp(colorChannel, "blue") == 0 ) { + for ( i = 0; i < mlData->samplesCount - 1; i++ ) { + fprintf ( fp, "%d\n", image->data[i].blue ); } - - } - else if (strcmp(colorChannel, "blue") == 0) { - for (i = 0; i < mlData->samplesCount - 1; i++) { - fprintf(fp, "%d\n", image->data[i].blue); - } - } - else { + } else { printf("Colorchannels are red, green and blue. Pick one of them!"); - exit(EXIT_FAILURE); + exit( EXIT_FAILURE ); } } fclose(fp); @@ -826,12 +831,12 @@ creating the SVG graph ====================================================================================================== */ -void colorSamples(FILE* fp, mldata_t *mlData) { +void colorSamples ( FILE* fp, mldata_t *mlData ) { int i = 0; - char *buffer = (char *)malloc(sizeof(char) * mlData->samplesCount + 1); + char *buffer = (char *) malloc(sizeof(char) * mlData->samplesCount + 1); while (!feof(fp)) { - if (fgets(buffer, mlData->samplesCount, fp) != NULL) { + if (fgets(buffer, mlData->samplesCount, fp) != NULL) { sscanf(buffer, "%lf", &xSamples[i]); //printf("%lf\n", xSamples[i] ); points[i].yVal[0] = xSamples[i]; // Fills points so actual input values can be seen as a graph @@ -847,7 +852,7 @@ void colorSamples(FILE* fp, mldata_t *mlData) { windowXMean -returns mean value of given input +returns mean value of given input ====================================================================================================== */ @@ -856,7 +861,7 @@ double windowXMean(int _arraylength, int xCount) { double *ptr; for (ptr = &xSamples[xCount - _arraylength]; ptr != &xSamples[xCount]; ptr++) { // Set ptr to beginning of window - sum += *ptr; + sum += *ptr; } return sum / (double)_arraylength; } @@ -864,43 +869,43 @@ double windowXMean(int _arraylength, int xCount) { /* ====================================================================================================== -usage - -used in conjunction with the args parser. Returns help section of "-h" + usage + + used in conjunction with the args parser. Returns help section of "-h" ====================================================================================================== */ -void usage(char **argv) { +void usage ( char **argv ) { printf("Usage: %s [POSIX style options] -i file ...\n", &argv[0][0]); printf("POSIX options:\n"); printf("\t-h\t\t\tDisplay this information.\n"); printf("\t-i \t\tName of inputfile. Must be PPM image.\n"); - printf("\t-n \t\tAmount of input data used.\n"); + printf("\t-n \t\tAmount of input data used.\n"); printf("\t-c \t\tUse this color channel from inputfile.\n"); printf("\t-w \t\tCount of used weights (windowSize).\n"); - printf("\t-l \t\tLearnrate, 0 < learnrate < 1.\n"); + printf("\t-l \t\tLearnrate, 0 < learnrate < 1.\n"); printf("\t-x true\t\t\tLogfiles only, no graph building.\n\t\t\t\tChoose for intense amount of input data.\n"); - printf("\t-s \t\tDigit for random seed generator.\n\t\t\t\tSame Digits produce same random values. Default is srand by time.\n"); + printf("\t-s \t\tDigit for random seed generator.\n\t\t\t\tSame Digits produce same random values. Default is srand by time.\n"); printf("\n\n"); - printf("%s compares prediction methods of least mean square filters.\nBy default it reads ppm file format and return logfiles as well\nas an svg graphs as an output of said least mean square methods.\n\nExample:\n\t%s -i myimage.ppm -w 3 -c green -s 5 -x true\n", &argv[0][0], &argv[0][0]); + printf("%s compares prediction methods of least mean square filters.\nBy default it reads ppm file format and return logfiles as well\nas an svg graphs as an output of said least mean square methods.\n\nExample:\n\t%s -i myimage.ppm -w 3 -c green -s 5 -x true\n", &argv[0][0], &argv[0][0]); exit(8); } /* ====================================================================================================== -init_mldata_t - - -Contains meachine learning data + init_mldata_t + + + Contains meachine learning data ====================================================================================================== */ mldata_t * init_mldata_t(unsigned windowSize, unsigned samplesCount, double learnrate) { - mlData = (mldata_t *)malloc(sizeof(mldata_t)); + mlData = (mldata_t *) malloc( sizeof(mldata_t) ); mlData->windowSize = windowSize; mlData->samplesCount = samplesCount; mlData->learnrate = learnrate; - mlData->weights = (double *)malloc(sizeof(double) * windowSize + 1); + mlData->weights = (double *) malloc ( sizeof(double) * windowSize + 1 ); return mlData; } diff --git a/src/cpp_implementation/README.md b/src/cpp_implementation/README.md index 399fd5d..3b76662 100644 --- a/src/cpp_implementation/README.md +++ b/src/cpp_implementation/README.md @@ -23,11 +23,11 @@ There are a bunch of options you can predefine but do not have to. The only para | Parameter | Description | StdVal | |:----------|:-----------------------------:|:-------| -| -i | The inputfile, has to be PPM | none | -| -n | Amount of input data used | 500 | -| -w | Size of M (window) | 5 | +| -i | The inputfile, has to be PPM. | none | +| -n | Amount of input data used. | 500 | +| -w | Size of M (window). | 5 | | -c | Choose RGB color channel, green has least noise. | green | -| -l | Learnrate of machine learning | 0.4 | -| -x | Exclude graph building. Logfiles only, choose for insane amount of input data. 10Mio. Pixels tested so far.| none| +| -l | Learnrate of machine learning. | 0.4 | +| -g true | include graph building. Choose for amount of input data lower than 1200.| none| | -s | Seed randomizing weights. Choose for repoducability. | time(NULL)|