max_pool1d_backward_f64

void max_pool1d_backward_f64(const double * g, const int * argmax, double * dx, int B, int C, int L, int OL)

Double-precision counterpart to max_pool1d_backward_f32.

gconst double*

Upstream gradient of shape (B, C, OL).

argmaxconst int32_t*

Argmax indices from the matching forward call.

dxdouble*

Output gradient of shape (B, C, L); accumulated.

B, C, L, OLint

Layout extents.