Home
last modified time | relevance | path

Searched refs:max_memory_allocated (Results 1 – 17 of 17) sorted by relevance

/aosp_15_r20/external/pytorch/docs/source/
H A Dxpu.rst60 max_memory_allocated
H A Dcuda.rst109 max_memory_allocated
/aosp_15_r20/external/pytorch/torch/xpu/
H A Dmemory.py136 def max_memory_allocated(device: _device_t = None) -> int: function
H A D__init__.py441 max_memory_allocated,
/aosp_15_r20/external/pytorch/test/inductor/
H A Dtest_scatter_optimization.py195 peak_mem = torch.cuda.max_memory_allocated() / 10**9
/aosp_15_r20/external/pytorch/torch/distributed/tensor/examples/
H A Dconvnext_example.py244 max_allocated = torch.cuda.max_memory_allocated()
/aosp_15_r20/external/pytorch/docs/source/notes/
H A Dhip.rst91 :meth:`~torch.cuda.max_memory_allocated` to monitor memory occupied by
H A Dcuda.rst404 :meth:`~torch.cuda.max_memory_allocated` to monitor memory occupied by
/aosp_15_r20/external/pytorch/test/distributed/fsdp/
H A Dtest_checkpoint_wrapper.py192 return torch.cuda.max_memory_allocated()
/aosp_15_r20/external/pytorch/torch/profiler/
H A D_memory_profiler.py1164 max_memory_allocated = torch.cuda.max_memory_allocated(device)
/aosp_15_r20/external/pytorch/torch/cuda/
H A Dmemory.py416 def max_memory_allocated(device: Union[Device, int] = None) -> int: function
/aosp_15_r20/external/pytorch/test/
H A Dtest_cuda_multigpu.py136 max_m_arr = [torch.cuda.max_memory_allocated(device)]
154 new_max_m = torch.cuda.max_memory_allocated(device)
206 self.assertEqual(torch.cuda.max_memory_allocated(device), last_m_arr[0])
H A Dtest_optim.py953 max_mems.append(torch.cuda.max_memory_allocated())
H A Dtest_autograd.py6834 mem_no_checkpoint = torch.cuda.max_memory_allocated()
6839 mem_reentrant_checkpoint = torch.cuda.max_memory_allocated()
6844 mem_no_reentrant_checkpoint = torch.cuda.max_memory_allocated()
H A Dtest_cuda.py3117 val = torch.cuda.max_memory_allocated()
/aosp_15_r20/external/pytorch/test/distributed/
H A Dtest_c10d_functional_native.py253 mem_usage[i] = torch.cuda.max_memory_allocated()
/aosp_15_r20/external/pytorch/benchmarks/dynamo/
H A Dcommon.py1963 return torch.cuda.max_memory_allocated() / 10**9