otx.algo.strategies#

Lightning strategy for single XPU device.

Classes

SingleXPUStrategy([device, accelerator, ...])

Strategy for training on single XPU device.

class otx.algo.strategies.SingleXPUStrategy(device: _DEVICE = 'xpu:0', accelerator: pl.accelerators.Accelerator | None = None, checkpoint_io: CheckpointIO | None = None, precision_plugin: PrecisionPlugin | None = None)[source]#

Bases: SingleDeviceStrategy

Strategy for training on single XPU device.

setup_optimizers(trainer: pl.Trainer) → None[source]#: Sets up optimizers.

property is_distributed: bool#: Returns true if the strategy supports distributed training.