write_kernel_cache_to_file#

nvmath.bindings.cutensor.write_kernel_cache_to_file(intptr_t handle, filename)[source]#

Writes the –per library– kernel cache to file.

Parameters:

handle (intptr_t) – Opaque handle holding cuTENSOR’s library context.
filename (str) – Specifies the filename (including the absolute path) to the file that should hold all the cache information. Warning: an existing file will be overwritten.