otx.algo.accelerators
Lightning accelerator for XPU device.
Classes
XPUAccelerator()
|
Support for a XPU, optimized for large-scale machine learning. |
-
class otx.algo.accelerators.XPUAccelerator[source]
Bases: Accelerator
Support for a XPU, optimized for large-scale machine learning.
-
static auto_device_count() → int[source]
Returns number of XPU devices available.
-
get_device_stats(device: str | device) → dict[str, Any][source]
Returns XPU devices stats.
-
static get_parallel_devices(devices: list) → list[device][source]
Generates a list of parrallel devices.
-
static is_available() → bool[source]
Checks if XPU available.
-
static parse_devices(devices: str | list | device) → list[source]
Parses devices for multi-GPU training.
-
setup_device(device: device) → None[source]
Sets up the specified device.
-
teardown() → None[source]
Clean up any state created by the accelerator.