RandomFewShot

RandomFewShot

Bases: Optimizer

Select randomly the best predictions to populate the LM's prompt.

Parameters:

Name	Type	Description	Default
k	int	The number of examples to select (default 3) among the best predictions.	3
k_best	int	The max number of best predictions to select from (default 10).	10

Returns:

Type	Description
Optimizer	The created optimizer.

Source code in synalinks/src/optimizers/random_few_shot.py

@synalinks_export("synalinks.optimizers.RandomFewShot")
class RandomFewShot(Optimizer):
    """
    Select randomly the best predictions to populate the LM's prompt.

    Args:
        k (int): The number of examples to select (default 3) among the best predictions.
        k_best (int): The max number of best predictions to select from (default 10).

    Returns:
        (Optimizer): The created optimizer.
    """

    def __init__(
        self,
        k=3,
        k_best=10,
        name=None,
        description=None,
    ):
        super().__init__(
            name=name,
            description=description,
            data_model=FewShotOptimizedVariable,
        )
        self.k = k
        self.k_best = k_best

    def build(self, variables):
        self.built = True

    async def optimize(self, trainable_variable, reward=None):
        """Perform a backprop/optimization on a single variable."""
        # Reward backpropagation
        predictions = trainable_variable.json().get("predictions")
        predictions = backpropagate_reward_to_predictions(predictions, reward)
        trainable_variable.json().update({"predictions": predictions})
        # Get the k best predictions (sorted by reward)
        sorted_predictions = sorted(
            predictions,
            key=lambda x: x[2] if x[2] is not None else float("-inf"),
            reverse=True,
        )
        top_k_predictions = sorted_predictions[: self.k_best]
        if len(top_k_predictions) > self.k:
            selected_predictions = random.sample(top_k_predictions, self.k)
        else:
            selected_predictions = top_k_predictions
        trainable_variable.json().update({"examples": selected_predictions})

    async def finalize(self, trainable_variable):
        """Finalize the optimization of a single variable (cleanup/scaling etc.)."""
        trainable_variable.json().update({"predictions": []})

    def get_config(self):
        return {
            "k": self.k,
            "k_best": self.k_best,
            "name": self.name,
            "description": self.description,
        }

finalize(trainable_variable) async

Finalize the optimization of a single variable (cleanup/scaling etc.).

Source code in synalinks/src/optimizers/random_few_shot.py

async def finalize(self, trainable_variable):
    """Finalize the optimization of a single variable (cleanup/scaling etc.)."""
    trainable_variable.json().update({"predictions": []})

optimize(trainable_variable, reward=None) async

Perform a backprop/optimization on a single variable.

Source code in synalinks/src/optimizers/random_few_shot.py

async def optimize(self, trainable_variable, reward=None):
    """Perform a backprop/optimization on a single variable."""
    # Reward backpropagation
    predictions = trainable_variable.json().get("predictions")
    predictions = backpropagate_reward_to_predictions(predictions, reward)
    trainable_variable.json().update({"predictions": predictions})
    # Get the k best predictions (sorted by reward)
    sorted_predictions = sorted(
        predictions,
        key=lambda x: x[2] if x[2] is not None else float("-inf"),
        reverse=True,
    )
    top_k_predictions = sorted_predictions[: self.k_best]
    if len(top_k_predictions) > self.k:
        selected_predictions = random.sample(top_k_predictions, self.k)
    else:
        selected_predictions = top_k_predictions
    trainable_variable.json().update({"examples": selected_predictions})