Intel MKL函数之 cblas_sgemm、cblas_sgemm_batch

cblas_sgemm

        int m = 40;
        int k = 20;
        int n = 40;
        std::vector<float> a(m*k, 1.0);
        std::vector<float> b(k*n, 1.0);
        std::vector<float> c(m*n, 0.0);

        float alpha = 1.0;
        float beta = 0.0;
        cblas_sgemm(CblasRowMajor, CblasNoTrans, CblasNoTrans, 
                    m, n, k, alpha,
                    a.data(), k, 
                    b.data(), n, beta,
                    c.data(), n);

        std::cout << "a.size(): " << a.size() << std::endl;
        for (int i = 0; i < a.size(); ++i) {
            std::cout << a[i] << " ";
            if ((i + 1) % k == 0)
                std::cout << std::endl;
        }

        std::cout << "b.size(): " << b.size() << std::endl;
        for (int i = 0; i < b.size(); ++i) {
            std::cout << b[i] << " ";
            if ((i + 1) % n == 0)
                std::cout << std::endl;
        }

        std::cout << "c.size(): " << c.size() << std::endl;
        for (int i = 0; i < c.size(); ++i) {
            std::cout << c[i] << " ";
            if ((i + 1) % n == 0)
                std::cout << std::endl;