c++ 无法在单层感知器网络中找到错误

des4xlb0  于 2023-05-02  发布在  其他
关注(0)|答案(1)|浏览(94)

我正在实现一个单层感知器,在代码中找不到问题。问题是,当我运行一个程序时,我的第二个数据集是X(0,0,1),权重系数是W(0,0,0.01),因此我的激活应该等于0。01。虽然我得到的输出为0。下面是我的代码:

#include <iostream>
#include <vector>
#include <cstdlib>
#include <ctime>

using namespace std;

//Helper function to generate a random number between 0 and 1
double weight_coef() {
    //((double)rand() / RAND_MAX);
    srand(time(NULL));
    double random = static_cast<double>(rand() / RAND_MAX);
    random = round(random * 100) / 100.0; //to 2 dec points
    return random;
}

// Function to calculate the dot product of two vectors
double activation(vector<int> x, vector<double> w) {
    double result = 0.0;
    for (int i = 0; i < x.size(); i++) {
        result += static_cast<double>(x[i] * w[i]);
    }
    return result;
}

int main() {
    // Set up the data
    vector<vector<int>> X = { {0, 0, 0}, {0, 0, 1}, {0, 1, 0}, {0, 1, 1},
                             {1, 0, 0}, {1, 0, 1}, {1, 1, 0}, {1, 1, 1} };
    vector<int> T = { 0, 1, 1, 1, 0, 1, 1, 1 };

    // Set up the variables
    vector<double> w = { weight_coef(), weight_coef(), weight_coef()};
    double theta = 0.5;
    //double nu = 0.3;
    double nu = 0.01;
    //double nu = 0.02;
    int num_iterations = 0;
    bool all_errors_zero = false;

    // Loop until all errors are zero for all data points
    while (!all_errors_zero) {
        all_errors_zero = true;
        num_iterations++;

        // Loop through all data points
        for (int i = 0; i < X.size(); i++) {
            // Calculate the sum and output Y
            double sum = activation(X[i], w);
            int Y = (sum > theta) ? 1 : 0;

            // Calculate the error and delta
            double error = nu * (T[i] - Y);
            vector<double> delta = { 
                error * X[i][0],
                error * X[i][1],
                error * X[i][2]};

            double deltaTheta = error * theta;

            // Update the weights
            for (int j = 0; j < w.size(); j++) {
                w[j] += delta[j];
            }

            // Check if the error is zero
            if (error != 0) {
                all_errors_zero = false;
            }

            // Output the results
            cout << "Weights: [" << w[0] << ", " << w[1] << ", " << w[2] << "]\n";
            cout << "Data set " << i + 1 << ":\n";
            cout << "Activation: " << sum << "\n";
            cout << "Y: " << Y << "\n";
            cout << "T: " << T[i] << "\n";
            cout << "Delta: " << error << "\n";
            cout << "Delta * Weights: [" << delta[0] << ", " << delta[1] << ", " << delta[2] << "]\n";
            cout << "Delta * Theta: " << deltaTheta << "\n\n";
        }
    }

    // Output the final results
    cout << "Final weights: [" << w[0] << ", " << w[1] << ", " << w[2] << "]\n";
    cout << "Number of iterations: " << num_iterations << "\n";

    return 0;
}

如果你能告诉我出了什么问题,我会很感激的。

iqxoj9l9

iqxoj9l91#

问题是std::rand返回一个保证在0和RAND_MAX之间的整数。因此权重系数代码

double weight_coef() {
    srand(time(NULL));
    double random = static_cast<double>(rand() / RAND_MAX); // <-- LOOK
    /* The above line will always set random to 0 */
    random = round(random * 100) / 100.0;
    return random;
}

将始终返回0(实际上它将 * 几乎 * 始终为0,并且将以概率1/RAND_MAX为1)。一个解决方案是取随机整数,将其除以100,然后将其解释为两位小数的概率。但请注意,这将是稍微不均匀的,因为RAND_MAX可能不能被100整除(可能是2^31 - 1 = 2147483647)。
如果你想制造一个均匀随机变量,你可以做类似的事情

#include <iostream>
#include <random>

const int LOWER_BOUND = 0;
const int UPPER_BOUND = 100;

std::random_device rand_dev;
std::mt19937 gen;
std::uniform_int_distribution<int> dist;

int random_int() {
  return dist(gen);
}

int main()
{
  gen = std::mt19937(rand_dev());
  dist = std::uniform_int_distribution<int>(LOWER_BOUND, UPPER_BOUND);

  for (int i = 0; i < 10; i++) {
      std::cout << random_int() << "\n";   // generate 10 random integers
  }
}

我注意到在每次函数调用时重复地重新初始化种子也不是一个好主意。相反,设置种子一次--可能是在main中(就像我发布的代码片段中隐式地完成的那样)。

相关问题