pytorch c++ 切片

0，nunpy切片

>>> import numpy as np
>>> a = np.arange(24).reshape(2,3,4)
>>> a
array([[[ 0,  1,  2,  3],
        [ 4,  5,  6,  7],
        [ 8,  9, 10, 11]],

       [[12, 13, 14, 15],
        [16, 17, 18, 19],
        [20, 21, 22, 23]]])
>>> a[:-1,:]
array([[[ 0,  1,  2,  3],
        [ 4,  5,  6,  7],
        [ 8,  9, 10, 11]]])
>>>

1，pytorch 切片

#include <torch/script.h>
using namespace std;
using namespace at;
using namespace torch;

int main(int argc, const char* argv[])
{ 
    torch::manual_seed(0);
    // auto a = torch::randn({4,4,4});
    auto a = at::arange(24).view({2,3,4});
    cout << "a size is " << a.sizes() << endl;
    cout << a << endl;
    // cout << a.size(0) << endl;
    // cout << a.size(1) << endl；

    auto b = at::slice(a,1,-1,0);                        # 这行有问题，两周内修改，其他事情
    													 # pytorch c++切片，一次只能切割一个维度，与python非常不一样
    cout << "b size is " << b.sizes() << endl;
    cout << b << endl;

    cout<< "ok\n";
    return 1;
}