关于matlab和python谁快的问题，python比matlab在乘法上快10倍，指数计算快4倍，加减运算持平，略慢于matlab。或许matlab只适合求解特征值。

python">import torch
import timen = 50000  # 矩阵规模
M = torch.rand(n, 31)start_time = time.time()
F_M = torch.exp(M)  # 将矩阵 M 映射到其指数值
end_time = time.time()print(f"Time taken: {end_time - start_time:.4f} seconds")n = 50000; % 矩阵规模
M = rand(n, 31);tic;
F_M = exp(M); % 将矩阵 M 映射到其指数值
toc;n = 50000; % 矩阵规模
M = rand(n, 31);tic;
F_M = M * M; % 将矩阵 M 映射到其指数值
toc;import torch
import timen = 50000  # 矩阵规模
M = torch.rand(n, 31)start_time = time.time()
F_M = M * M  # 将矩阵 M 映射到其指数值
end_time = time.time()print(f"Time taken: {end_time - start_time:.4f} seconds")