Index

A

AcceleratedOptimizer (class in accelerate.optimizer) Accelerator (class in accelerate)

AcceleratorState (class in accelerate.state) autocast() (accelerate.Accelerator method)

B

backward() (accelerate.Accelerator method)

BatchSamplerShard (class in accelerate.data_loader)

C

clip_grad_norm_() (accelerate.Accelerator method)

clip_grad_value_() (accelerate.Accelerator method)

D

DataLoaderShard (class in accelerate.data_loader)

DistributedDataParallelKwargs (class in accelerate) DistributedType (class in accelerate.state)

E

extract_model_from_parallel() (in module accelerate.utils)

F

free_memory() (accelerate.Accelerator method)

G

gather() (accelerate.Accelerator method) (in module accelerate.utils)

GradScalerKwargs (class in accelerate)

I

is_local_main_process() (accelerate.Accelerator property)

is_main_process() (accelerate.Accelerator property) IterableDatasetShard (class in accelerate.data_loader)

L

local_main_process_first() (accelerate.Accelerator method)

M

main_process_first() (accelerate.Accelerator method)

N

notebook_launcher() (in module accelerate)

O

optimizer_step_was_skipped() (accelerate.Accelerator property)

P

pad_across_processes() (accelerate.Accelerator method) prepare() (accelerate.Accelerator method)

prepare_data_loader() (in module accelerate.data_loader) print() (accelerate.Accelerator method)

S

save() (accelerate.Accelerator method) send_to_device() (in module accelerate.utils)

set_seed() (in module accelerate.utils) synchronize_rng_state() (in module accelerate.utils) synchronize_rng_states() (in module accelerate.utils)

U

unscale_gradients() (accelerate.Accelerator method)

unwrap_model() (accelerate.Accelerator method)

W

wait_for_everyone() (accelerate.Accelerator method) (in module accelerate.utils)