# BEMA for Reference Model

This feature implements the BEMA algorithm to update the reference model during DPO training.

## Usage

```python
from trl.experimental.bema_for_ref_model import BEMACallback, DPOTrainer
from datasets import load_dataset

dataset = load_dataset("trl-internal-testing/zen", "standard_preference", split="train")

bema_callback = BEMACallback(update_ref_model=True)

trainer = DPOTrainer(
    model="trl-internal-testing/tiny-Qwen2ForCausalLM-2.5",
    train_dataset=dataset,
    callbacks=[bema_callback],
)
trainer.train()
```

## DPOTrainer[[trl.DPOTrainer]]

#### trl.DPOTrainer[[trl.DPOTrainer]]

[Source](https://github.com/huggingface/trl/blob/v1.1.0/trl/experimental/bema_for_ref_model/dpo_trainer.py#L19)

traintrl.DPOTrainer.trainhttps://github.com/huggingface/trl/blob/v1.1.0/transformers/trainer.py#L1322[{"name": "resume_from_checkpoint", "val": ": str | bool | None = None"}, {"name": "trial", "val": ": optuna.Trial | dict[str, Any] | None = None"}, {"name": "ignore_keys_for_eval", "val": ": list[str] | None = None"}]- **resume_from_checkpoint** (`str` or `bool`, *optional*) --
  If a `str`, local path to a saved checkpoint as saved by a previous instance of `Trainer`. If a
  `bool` and equals `True`, load the last checkpoint in *args.output_dir* as saved by a previous instance
  of `Trainer`. If present, training will resume from the model/optimizer/scheduler states loaded here.
- **trial** (`optuna.Trial` or `dict[str, Any]`, *optional*) --
  The trial run or the hyperparameter dictionary for hyperparameter search.
- **ignore_keys_for_eval** (`list[str]`, *optional*) --
  A list of keys in the output of your model (if it is a dictionary) that should be ignored when
  gathering predictions for evaluation during the training.0`~trainer_utils.TrainOutput`Object containing the global step count, training loss, and metrics.

Main training entry point.

**Parameters:**

resume_from_checkpoint (`str` or `bool`, *optional*) : If a `str`, local path to a saved checkpoint as saved by a previous instance of `Trainer`. If a `bool` and equals `True`, load the last checkpoint in *args.output_dir* as saved by a previous instance of `Trainer`. If present, training will resume from the model/optimizer/scheduler states loaded here.

trial (`optuna.Trial` or `dict[str, Any]`, *optional*) : The trial run or the hyperparameter dictionary for hyperparameter search.

ignore_keys_for_eval (`list[str]`, *optional*) : A list of keys in the output of your model (if it is a dictionary) that should be ignored when gathering predictions for evaluation during the training.

**Returns:**

``~trainer_utils.TrainOutput``

Object containing the global step count, training loss, and metrics.
#### save_model[[trl.DPOTrainer.save_model]]

[Source](https://github.com/huggingface/trl/blob/v1.1.0/transformers/trainer.py#L3745)

Will save the model, so you can reload it using `from_pretrained()`.

Will only save from the main process.
#### push_to_hub[[trl.DPOTrainer.push_to_hub]]

[Source](https://github.com/huggingface/trl/blob/v1.1.0/transformers/trainer.py#L3992)

Upload `self.model` and `self.processing_class` to the 🤗 model hub on the repo `self.args.hub_model_id`.

**Parameters:**

commit_message (`str`, *optional*, defaults to `"End of training"`) : Message to commit while pushing.

blocking (`bool`, *optional*, defaults to `True`) : Whether the function should return only when the `git push` has finished.

token (`str`, *optional*, defaults to `None`) : Token with write permission to overwrite Trainer's original args.

revision (`str`, *optional*) : The git revision to commit from. Defaults to the head of the "main" branch.

kwargs (`dict[str, Any]`, *optional*) : Additional keyword arguments passed along to `~Trainer.create_model_card`.

**Returns:**

The URL of the repository where the model was pushed if `blocking=False`, or a `Future` object tracking the
progress of the commit if `blocking=True`.

## BEMACallback[[trl.BEMACallback]]

#### trl.BEMACallback[[trl.BEMACallback]]

[Source](https://github.com/huggingface/trl/blob/v1.1.0/trl/experimental/bema_for_ref_model/callback.py#L59)

A [TrainerCallback](https://huggingface.co/docs/transformers/v5.5.3/en/main_classes/callback#transformers.TrainerCallback) that implements [BEMA](https://huggingface.co/papers/2508.00180)
(Bias-Corrected Exponential Moving Average) by [Adam Block](https://huggingface.co/abblock) and [Cyril
Zhang](https://huggingface.co/cyrilzhang). Code from https://github.com/abblock/bema under MIT license.

BEMA computes model weights that scale like:

$$
\theta_t' = \alpha_t \cdot (\theta_t - \theta_0) + \text{EMA}_t
$$

where  \\( \theta_t \\) is the current model weights,  \\( \theta_0 \\) is a snapshot of the model weights at the
first `update_after` step,  \\( \text{EMA}_t  \\) is the exponential moving average of the model weights, and
\\( \alpha_t \\) is a scaling factor that decays with the number of steps  \\( t \\) as

$$
\alpha_t = (\rho + \gamma \cdot t)^{-\eta}.
$$

The EMA is computed as:

$$
\text{EMA}_t = (1 - \beta_t) \cdot \text{EMA}_{t-1} + \beta_t \cdot \theta_t
$$

where  \\( \beta_t \\) is a decay factor that decays with the number of steps  \\( t \\) as

$$
\beta_t = (\rho + \gamma \cdot t)^{-\kappa}.
$$

Example:

```python
from trl import BEMACallback

trainer = Trainer(..., callbacks=[BEMACallback()])
```

**Parameters:**

update_freq (`int`, *optional*, defaults to `400`) : Update the BEMA weights every X steps. Denoted this as  \\( \phi \\) in the paper.

ema_power (`float`, *optional*, defaults to `0.5`) : Power for the EMA decay factor. Denoted  \\( \kappa \\) in the paper. To disable EMA, set this to `0.0`.

bias_power (`float`, *optional*, defaults to `0.2`) : Power for the BEMA scaling factor. Denoted  \\( \eta \\) in the paper. To disable BEMA, set this to `0.0`.

lag (`int`, *optional*, defaults to `10`) : Initial offset in the weight decay schedule that controls early-stage smoothness by acting as a virtual starting age for the updates. Denoted as  \\( \rho \\) in the paper.

update_after (`int`, *optional*, defaults to `0`) : Burn-in time before starting to update the BEMA weights. Denoted  \\( \tau \\) in the paper.

multiplier (`float`, *optional*, defaults to `1.0`) : Initial value for the EMA decay factor. Denoted as  \\( \gamma \\) in the paper.

min_ema_multiplier (`float`, *optional*, defaults to `0.0`) : Minimum value for the EMA decay factor.

device (`str`, *optional*, defaults to `"cpu"`) : Device to use for the BEMA buffers, e.g. `"cpu"` or `"cuda"`. Note that in most cases, this device SHOULD BE DIFFERENT from the device used for training in order to avoid OOM.

update_ref_model (`bool`, *optional*, defaults to `False`) : Whether to update the reference model with BEMA weights. This creates a lagged, smoothed version of the main model as the reference model.

ref_model_update_freq (`int`, *optional*, defaults to `400`) : Update the reference model with BEMA weights every this many steps.

ref_model_update_after (`int`, *optional*, defaults to `0`) : Number of steps to wait before starting to update the reference model.