# DeepSpeed utilities

## DeepSpeedPlugin

## get_active_deepspeed_plugin[[accelerate.utils.get_active_deepspeed_plugin]]

#### accelerate.utils.get_active_deepspeed_plugin[[accelerate.utils.get_active_deepspeed_plugin]]

```python
accelerate.utils.get_active_deepspeed_plugin(state)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L100)

**Raises:** ``ValueError``

- ``ValueError`` -- If DeepSpeed was not enabled and this function is called.

Returns the currently active DeepSpeedPlugin.

#### accelerate.DeepSpeedPlugin[[accelerate.DeepSpeedPlugin]]

```python
accelerate.DeepSpeedPlugin(hf_ds_config: typing.Any = None, gradient_accumulation_steps: int = None, gradient_clipping: float = None, zero_stage: int = None, is_train_batch_min: bool = True, offload_optimizer_device: str = None, offload_param_device: str = None, offload_optimizer_nvme_path: str = None, offload_param_nvme_path: str = None, zero3_init_flag: bool = None, zero3_save_16bit_model: bool = None, transformer_moe_cls_names: str = None, enable_msamp: bool = None, msamp_opt_level: typing.Optional[typing.Literal['O1', 'O2']] = None)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/dataclasses.py#L1122)

**Parameters:**

hf_ds_config (`Any`, defaults to `None`) : Path to DeepSpeed config file or dict or an object of class `accelerate.utils.deepspeed.HfDeepSpeedConfig`.

gradient_accumulation_steps (`int`, defaults to `None`) : Number of steps to accumulate gradients before updating optimizer states. If not set, will use the value from the `Accelerator` directly.

gradient_clipping (`float`, defaults to `None`) : Enable gradient clipping with value.

zero_stage (`int`, defaults to `None`) : Possible options are 0, 1, 2, 3. Default will be taken from environment variable.

is_train_batch_min (`bool`, defaults to `True`) : If both train & eval dataloaders are specified, this will decide the `train_batch_size`.

offload_optimizer_device (`str`, defaults to `None`) : Possible options are none|cpu|nvme. Only applicable with ZeRO Stages 2 and 3.

offload_param_device (`str`, defaults to `None`) : Possible options are none|cpu|nvme. Only applicable with ZeRO Stage 3.

offload_optimizer_nvme_path (`str`, defaults to `None`) : Possible options are /nvme|/local_nvme. Only applicable with ZeRO Stage 3.

offload_param_nvme_path (`str`, defaults to `None`) : Possible options are /nvme|/local_nvme. Only applicable with ZeRO Stage 3.

zero3_init_flag (`bool`, defaults to `None`) : Flag to indicate whether to save 16-bit model. Only applicable with ZeRO Stage-3.

zero3_save_16bit_model (`bool`, defaults to `None`) : Flag to indicate whether to save 16-bit model. Only applicable with ZeRO Stage-3.

transformer_moe_cls_names (`str`, defaults to `None`) : Comma-separated list of Transformers MoE layer class names (case-sensitive). For example, `MixtralSparseMoeBlock`, `Qwen2MoeSparseMoeBlock`, `JetMoEAttention`, `JetMoEBlock`, etc.

enable_msamp (`bool`, defaults to `None`) : Flag to indicate whether to enable MS-AMP backend for FP8 training.

msamp_opt_level (`Optional[Literal["O1", "O2"]]`, defaults to `None`) : Optimization level for MS-AMP (defaults to 'O1'). Only applicable if `enable_msamp` is True. Should be one of ['O1' or 'O2'].

This plugin is used to integrate DeepSpeed.

#### deepspeed_config_process[[accelerate.DeepSpeedPlugin.deepspeed_config_process]]

```python
deepspeed_config_process(prefix = '', mismatches = None, config = None, must_match = True, **kwargs)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/dataclasses.py#L1392)

Process the DeepSpeed config with the values from the kwargs.

#### select[[accelerate.DeepSpeedPlugin.select]]

```python
select(_from_accelerator_state: bool = False)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/dataclasses.py#L1554)

Sets the HfDeepSpeedWeakref to use the current deepspeed plugin configuration

#### accelerate.utils.DummyScheduler[[accelerate.utils.DummyScheduler]]

```python
accelerate.utils.DummyScheduler(optimizer, total_num_steps = None, warmup_num_steps = 0, lr_scheduler_callable = None, **kwargs)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L362)

**Parameters:**

optimizer (`torch.optim.optimizer.Optimizer`) : The optimizer to wrap.

total_num_steps (int, *optional*) : Total number of steps.

warmup_num_steps (int, *optional*) : Number of steps for warmup.

lr_scheduler_callable (callable, *optional*) : A callable function that creates an LR Scheduler. It accepts only one argument `optimizer`.

- ****kwargs** (additional keyword arguments, *optional*) : Other arguments.

Dummy scheduler presents model parameters or param groups, this is primarily used to follow conventional training
loop when scheduler config is specified in the deepspeed config file.

## DeepSpeedEnginerWrapper[[accelerate.utils.DeepSpeedEngineWrapper]]

#### accelerate.utils.DeepSpeedEngineWrapper[[accelerate.utils.DeepSpeedEngineWrapper]]

```python
accelerate.utils.DeepSpeedEngineWrapper(engine)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L253)

**Parameters:**

engine (deepspeed.runtime.engine.DeepSpeedEngine) : deepspeed engine to wrap

Internal wrapper for deepspeed.runtime.engine.DeepSpeedEngine. This is used to follow conventional training loop.

#### get_global_grad_norm[[accelerate.utils.DeepSpeedEngineWrapper.get_global_grad_norm]]

```python
get_global_grad_norm()
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L286)

Get the global gradient norm from DeepSpeed engine.

## DeepSpeedOptimizerWrapper[[accelerate.utils.DeepSpeedOptimizerWrapper]]

#### accelerate.utils.DeepSpeedOptimizerWrapper[[accelerate.utils.DeepSpeedOptimizerWrapper]]

```python
accelerate.utils.DeepSpeedOptimizerWrapper(optimizer)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L295)

**Parameters:**

optimizer (`torch.optim.optimizer.Optimizer`) : The optimizer to wrap.

Internal wrapper around a deepspeed optimizer.

## DeepSpeedSchedulerWrapper[[accelerate.utils.DeepSpeedSchedulerWrapper]]

#### accelerate.utils.DeepSpeedSchedulerWrapper[[accelerate.utils.DeepSpeedSchedulerWrapper]]

```python
accelerate.utils.DeepSpeedSchedulerWrapper(scheduler, optimizers)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L322)

**Parameters:**

scheduler (`torch.optim.lr_scheduler.LambdaLR`) : The scheduler to wrap.

optimizers (one or a list of `torch.optim.Optimizer`) --

Internal wrapper around a deepspeed scheduler.

## DummyOptim[[accelerate.utils.DummyOptim]]

#### accelerate.utils.DummyOptim[[accelerate.utils.DummyOptim]]

```python
accelerate.utils.DummyOptim(params, lr = 0.001, weight_decay = 0, **kwargs)
```

[Source](https://github.com/huggingface/accelerate/blob/main/src/accelerate/utils/deepspeed.py#L339)

**Parameters:**

lr (float) : Learning rate.

params (iterable) : iterable of parameters to optimize or dicts defining parameter groups

weight_decay (float) : Weight decay.

- ****kwargs** (additional keyword arguments, *optional*) : Other arguments.

Dummy optimizer presents model parameters or param groups, this is primarily used to follow conventional training
loop when optimizer config is specified in the deepspeed config file.

## DummyScheduler