Exceptions and Warnings

General exceptions used by DPNP. Note that some exceptions may be module specific, such as linear algebra errors.

Exceptions

exception dpnp.exceptions.AxisError[source]: Given when an axis is invalid.

exception dpnp.exceptions.DLPackCreationError

Given when constructing DLPack capsule from either dpnp.ndarray or dpctl.tensor.usm_ndarray based on a USM allocation on a partitioned SYCL device.

Examples

>>> import dpnp as np
>>> import dpctl
>>> dev = dpctl.SyclDevice('cpu')
>>> sdevs = dev.create_sub_devices(partition=[1, 1])
>>> q = dpctl.SyclQueue(sdevs[0])
>>> x = np.ones(10, sycl_queue=q)
>>> np.from_dlpack(x)
Traceback (most recent call last):
...
DLPackCreationError: to_dlpack_capsule: DLPack can only export arrays based on USM allocations bound to a default platform SYCL context

exception dpnp.exceptions.ExecutionPlacementError: Given when execution placement target can not be unambiguously determined from input arrays. Make sure that input arrays are associated with the same dpctl.SyclQueue, or migrate data to the same dpctl.SyclQueue using dpnp.ndarray.to_device() method.

exception dpnp.exceptions.SyclContextCreationError: Given when dpctl.SyclContext instance could not be created.

exception dpnp.exceptions.SyclDeviceCreationError: Given when dpctl.SyclDevice instance could not be created.

exception dpnp.exceptions.SyclQueueCreationError: Given when dpctl.SyclQueue instance could not be created. The creation can fail if the filter string is invalid, or the backend or device type values are not supported.

exception dpnp.exceptions.USMAllocationError

Given when Unified Shared Memory (USM) allocation call returns a null pointer, signaling a failure to perform the allocation. Some common reasons for allocation failure are:

insufficient free memory to perform the allocation request

allocation size exceeds the maximum supported by targeted backend