How to Implement Your Callback with OptunaHub

This recipe shows how to implement and register your own callback with OptunaHub.

Callbacks are used when you want to insert custom processing after each trial completes. Typical use cases include:

• Uploading the current best result to an external server (e.g., W&B, MLflow).

• Sending a notification when a new best value is found.

• Stopping the study early based on custom criteria by calling stop().

A callback is simply a callable with the following signature:

def callback(study: optuna.Study, trial: optuna.trial.FrozenTrial) -> None:
    ...

Optuna calls each registered callback once per trial, after the objective function returns and the trial state has been recorded. At that point Study already reflects the updated best value / best params, and trial is a FrozenTrial (immutable).

The simplest way to implement a callback is to define a plain function:

from __future__ import annotations

import optuna


def my_callback(study: optuna.Study, trial: optuna.trial.FrozenTrial) -> None:
    print(f"Trial {trial.number} finished.")
    print(f"  params     : {trial.params}")
    print(f"  value      : {trial.value}")
    if trial.state == optuna.trial.TrialState.COMPLETE:
        print(f"  best_value : {study.best_value}")

If your callback needs to hold internal state (e.g., a connection to an external service), you can implement it as a class with a __call__ method instead.

class MyCallback:
    """A callback that prints trial information after each completed trial.

    Args:
        verbose: If ``True``, also print the full ``params`` dict.
    """

    def __init__(self, verbose: bool = True) -> None:
        self._verbose = verbose

    def __call__(self, study: optuna.Study, trial: optuna.trial.FrozenTrial) -> None:
        # This method is called after every trial regardless of its state.
        if trial.state != optuna.trial.TrialState.COMPLETE:
            return

        print(f"Trial {trial.number} finished.")
        if self._verbose:
            print(f"  params     : {trial.params}")
        print(f"  value      : {trial.value}")
        print(f"  best_value : {study.best_value}")

The callback is passed to optimize() via the callbacks argument. Multiple callbacks can be specified as a list; they are called in order after each trial.

def objective(trial: optuna.trial.Trial) -> float:
    x = trial.suggest_float("x", -10, 10)
    y = trial.suggest_int("y", -5, 5)
    return x**2 + y**2

Run the study with a plain function callback.

study = optuna.create_study()
study.optimize(objective, n_trials=5, callbacks=[my_callback])

Trial 0 finished.
  params     : {'x': -0.8284883529148104, 'y': -1}
  value      : 1.6863929509154953
  best_value : 1.6863929509154953
Trial 1 finished.
  params     : {'x': -3.00542174870907, 'y': 3}
  value      : 18.032559887613484
  best_value : 1.6863929509154953
Trial 2 finished.
  params     : {'x': 0.92369337241945, 'y': 0}
  value      : 0.8532094462516169
  best_value : 0.8532094462516169
Trial 3 finished.
  params     : {'x': -1.9198280195936963, 'y': -1}
  value      : 4.685739624817054
  best_value : 0.8532094462516169
Trial 4 finished.
  params     : {'x': 5.4744903608121565, 'y': -4}
  value      : 45.97004471062522
  best_value : 0.8532094462516169

Run another study with the class-based callback.

study = optuna.create_study()
study.optimize(objective, n_trials=5, callbacks=[MyCallback(verbose=True)])

Trial 0 finished.
  params     : {'x': 6.042968631575789, 'y': -1}
  value      : 37.517469882208964
  best_value : 37.517469882208964
Trial 1 finished.
  params     : {'x': -5.158391043660471, 'y': 0}
  value      : 26.60899815931656
  best_value : 26.60899815931656
Trial 2 finished.
  params     : {'x': -7.57249376635184, 'y': -4}
  value      : 73.34266184143749
  best_value : 26.60899815931656
Trial 3 finished.
  params     : {'x': 2.0268057515631988, 'y': 1}
  value      : 5.107941554569663
  best_value : 5.107941554569663
Trial 4 finished.
  params     : {'x': -9.608565092768808, 'y': -4}
  value      : 108.32452314197526
  best_value : 5.107941554569663

After implementing your own callback, you can register it with OptunaHub. See How to Register Your Package with OptunaHub for how to register your callback with OptunaHub. The category name to use when placing your package in the registry is callbacks: