torch.cuda.caching_allocator_delete

torch.cuda.caching_allocator_delete(mem_ptr) [source]

Deletes memory allocated using the CUDA memory allocator.

Memory allocated with caching_allocator_alloc(). is freed here. The associated device and stream are tracked inside the allocator.

Parameters:

mem_ptr (int) – memory address to be freed by the allocator.

Note

See Memory management for more details about GPU memory management.