remove comments, atoi to std::stoi

main.cpp

@@ -10,16 +10,24 @@
 #include "include/TimeSeriesPredictor.cuh"
 
 auto readDataToMemory() -> std::vector<float>;
+auto usage(char * arg) -> void;
 
 int main(int argc, char ** argv) {
+    int nodes, populationSize, windowSize;
+
     if(argc != 4) {
-        std::cerr << "Usage: " << argv[0] << " <num_of_nodes> <population_size> <window_size>" << std::endl;
+        usage(argv[0]);
         return 1;
     }
 
-    int nodes = atoi(argv[1]);
-    int populationSize = atoi(argv[2]);
-    int windowSize = atoi(argv[3]);
+    try {
+        nodes = std::stoi(argv[1]);
+        populationSize = std::stoi(argv[2]);
+        windowSize = std::stoi(argv[3]);
+    } catch(std::exception const & e) {
+        usage(argv[0]);
+        return 1;
+    }
 
     auto timeSeries = readDataToMemory();
 
@@ -28,6 +36,10 @@ int main(int argc, char ** argv) {
     return 0;
 }
 
+auto usage(char * arg) -> void {
+    std::cerr << "Usage: " << arg << " <num_of_nodes> <population_size> <window_size>" << std::endl;
+}
+
 auto readDataToMemory() -> std::vector<float> {
     std::string dataFolder = "./data/";
     std::vector<float> dataVector;

src/FitnessFunctionKernel.cu

@@ -18,11 +18,8 @@ __global__ void calculate_fitness(
     int i = 0; 
 
     while(i < data_size - window_size * node_number) {
-        // float * input_data = new float[window_size * node_number];
-        // memcpy(input_data, &data[i], window_size * node_number * sizeof(float));
 
         for(int j = 0; j < window_size * node_number; ++j) {
-            // map_input[j % node_number] += data_weights[j] * input_data[j];
             mapInput[j % node_number + index * node_number] += dataWeights[j + window_size * node_number * index] * data[i + j];
         }
 
@@ -31,8 +28,6 @@ __global__ void calculate_fitness(
         }
 
         for(int j = 0; j < node_number; ++j) {
-            // float * current_weights = new float[node_number];
-            // memcpy(current_weights, &mapWeights[j * node_number], node_number * sizeof(float));
 
             float x = 0.0f;
             for(int k = 0; k < node_number; ++k) {
@@ -42,10 +37,8 @@ __global__ void calculate_fitness(
             float y = 1.0 / (1.0 + expf(-5 * x));
             float prediction_error = abs(y - data[j+i+1]);
             cumulative_error += prediction_error;
-            // delete [] current_weights;
         }
         i += node_number;
-        // delete [] input_data;
     }
 
     float epsilon = cumulative_error / ((node_number * data_size) - (node_number * window_size));

src/TimeSeriesPredictor.cu

@@ -43,7 +43,7 @@ auto TimeSeriesPredictor::train() -> std::vector<float> {
         Chromosome bestCandidate = this->maxFitness(this->population);
         if(generation == 500 || abs(1.0 - bestCandidate.fitness) < 1e-5) break;
 
-        std::cout << "-----GEN {generation}-------" << std::endl;
+        std::cout << "-----GEN " << generation << " -------" << std::endl;
         std::cout << "Best fitness: " << bestCandidate.fitness << std::endl;
 
         while(nextGen.size() < this->populationSize) {
@@ -156,7 +156,7 @@ auto TimeSeriesPredictor::launchCudaKernel() -> void {
         gpuErrchk(cudaMemcpy(&this->mapWeightsGpu[i * n * n], &weights[w * n], n * n * sizeof(float), cudaMemcpyHostToDevice));
     }
 
-    cudaMemset(this->mapInputGpu, 0, n * sizeof(float));
+    gpuErrchk(cudaMemset(this->mapInputGpu, 0, n * sizeof(float)));
 
     calculate_fitness<<<4, 512>>>(this->dataWeightsGpu, this->mapWeightsGpu, this->dataGpu, this->fitnessGpu, this->mapInputGpu, w, n, this->populationSize, dataSize);
     gpuErrchk(cudaPeekAtLastError());