torch.cuda.memory.caching_allocator_delete# torch.cuda.memory.caching_allocator_delete(mem_ptr)[source]# Delete 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.