o i! @sddlmZddlZddlZddlmZddlmZ ddgZ Gdddej Z ej edddd d ed e dBd edBd dfddZdS))AnyN) _to_dlpack)Device DLDeviceType from_dlpackc@sHeZdZdZdZdZdZdZdZdZ dZ d Z d Z d Z d Zd ZdZdZdS)r))))))) ) ) ) ) ))))N)__name__ __module__ __qualname__kDLCPUkDLCUDA kDLCUDAHost kDLOpenCL kDLVulkankDLMetalkDLVPIkDLROCM kDLROCMHost kDLExtDevkDLCUDAManaged kDLOneAPI kDLWebGPU kDLHexagonkDLMAIAr(r(T/sda-disk/www/egybert/egybert_env/lib/python3.10/site-packages/torch/utils/dlpack.pyrs ato_dlpack(tensor) -> PyCapsule Returns an opaque object (a "DLPack capsule") representing the tensor. .. note:: ``to_dlpack`` is a legacy DLPack interface. The capsule it returns cannot be used for anything in Python other than use it as input to ``from_dlpack``. The more idiomatic use of DLPack is to call ``from_dlpack`` directly on the tensor object - this works when that object has a ``__dlpack__`` method, which PyTorch and most other libraries indeed have now. .. warning:: Only call ``from_dlpack`` once per capsule produced with ``to_dlpack``. Behavior when a capsule is consumed multiple times is undefined. Args: tensor: a tensor to be exported The DLPack capsule shares the tensor's memory. )devicecopy ext_tensorr*r+returnz torch.TensorcCsFt|dr i}d|d<|}d}d}|dur||d<|}|dur\t|tr,t|}t|tjs;tdt|tj |}||k}|sK||d <|r\|dur\t d |d |d |d t j t j fvrtjd|d} |d t j k} | r| jd krdn| j} | |d<d} z |jdi|} Wn tyYnw| dur|ddz |jdi|} Wn tyYnw| dur|ddd}z |jdi|} Wn tyYnw| dur|d d|jdi|} tj| } |dur|s|s| } |r | |} | S|dus|durtd|} tj| S)afrom_dlpack(ext_tensor) -> Tensor Converts a tensor from an external library into a ``torch.Tensor``. The returned PyTorch tensor will share the memory with the input tensor (which may have come from another library). Note that in-place operations will therefore also affect the data of the input tensor. This may lead to unexpected issues (e.g., other libraries may have read-only flags or immutable data structures), so the user should only do this if they know for sure that this is fine. Args: ext_tensor (object with ``__dlpack__`` attribute, or a DLPack capsule): The tensor or DLPack capsule to convert. If ``ext_tensor`` is a tensor (or ndarray) object, it must support the ``__dlpack__`` protocol (i.e., have a ``ext_tensor.__dlpack__`` method). Otherwise ``ext_tensor`` may be a DLPack capsule, which is an opaque ``PyCapsule`` instance, typically produced by a ``to_dlpack`` function or method. device (torch.device or str or None): An optional PyTorch device specifying where to place the new tensor. If None (default), the new tensor will be on the same device as ``ext_tensor``. copy (bool or None): An optional boolean indicating whether or not to copy ``self``. If None, PyTorch will copy only if necessary. Examples:: >>> import torch.utils.dlpack >>> t = torch.arange(4) # Convert a tensor directly (supported in PyTorch >= 1.10) >>> t2 = torch.from_dlpack(t) >>> t2[:2] = -1 # show that memory is shared >>> t2 tensor([-1, -1, 2, 3]) >>> t tensor([-1, -1, 2, 3]) # The old-style DLPack usage, with an intermediate capsule object >>> capsule = torch.utils.dlpack.to_dlpack(t) >>> capsule >>> t3 = torch.from_dlpack(capsule) >>> t3 tensor([-1, -1, 2, 3]) >>> t3[0] = -9 # now we're sharing memory between 3 tensors >>> t3 tensor([-9, -1, 2, 3]) >>> t2 tensor([-9, -1, 2, 3]) >>> t tensor([-9, -1, 2, 3]) __dlpack__)rr max_versionTFNr+z&from_dlpack: unsupported device type: dl_devicez&cannot move DLPack tensor from device z to z- without copying. Set copy=None or copy=True.rzcuda:rstreamzQdevice and copy kwargs not supported when ext_tensor is already a DLPack capsule.r()hasattr__dlpack_device__ isinstancestrtorchr*AssertionErrortype_C_torchDeviceToDLDevice ValueErrorrrr cudacurrent_stream cuda_streamr. TypeErrorpop _from_dlpackcloneto)r,r*r+kwargsrequested_copyproducer_handled_copycross_device_transfer ext_devicetarget_dl_devicer1is_cuda stream_ptrdlpacktensorr(r(r)r:s~ @              )typingrr6enumtorch._Cr to_dlpack torch.typesr_Device__all__IntEnumrr9 _add_docstrboolrr(r(r(r)s*