function_supported_devices#
- ivy.function_supported_devices(fn, recurse=True)[source]#
Return the supported devices of the current backend’s function. The function returns a dict containing the supported devices for the compositional and primary implementations in case of partial mixed functions.
- Parameters:
fn (
Callable
) – The function to check for the supported device attributerecurse (
bool
, default:True
) – Whether to recurse into used ivy functions. Default isTrue
.
- Return type:
Union
[Tuple
,dict
]- Returns:
ret – Tuple or dict containing the supported devices of the function
Examples
>>> import ivy >>> ivy.set_backend('numpy') >>> print(ivy.function_supported_devices(ivy.ones)) ('cpu',)
>>> ivy.set_backend('torch') >>> x = ivy.function_supported_devices(ivy.ones) >>> x = sorted(x) ('cpu', 'gpu')