optuna.yaml

# @package _global_

# example hyperparameter optimization of some experiment with Optuna:
# python train.py -m hparams_search=mnist_optuna experiment=example

defaults:
  - override /hydra/sweeper: optuna

# here we define Optuna hyperparameter search
# it optimizes for value returned from function with @hydra.main decorator
# docs: https://hydra.cc/docs/next/plugins/optuna_sweeper
hydra:
  mode: "MULTIRUN" # set hydra to multirun by default if this config is attached

  sweeper:
    _target_: hydra_plugins.hydra_optuna_sweeper.optuna_sweeper.OptunaSweeper

    # storage URL to persist optimization results
    # for example, you can use SQLite if you set 'sqlite:///example.db'
    storage: null

    # name of the study to persist optimization results
    study_name: null

    # number of parallel workers
    n_jobs: 1

    # 'minimize' or 'maximize' the objective
    direction: maximize

    # total number of runs that will be executed
    n_trials: 10

    # choose Optuna hyperparameter sampler
    # you can choose bayesian sampler (tpe), random search (without optimization), grid sampler, and others
    # docs: https://optuna.readthedocs.io/en/stable/reference/samplers.html
    sampler:
      _target_: optuna.samplers.TPESampler
      seed: 1234
      n_startup_trials: 10 # number of random sampling runs before optimization starts

    # define hyperparameter search space
    params:
      model/text_splitter: choice(RecursiveCharacter, CharacterTextSplitter) # if you want to change component level
      model.text_splitter.chunk_size: range(500, 1500, 100) # if you want to change specific hyperparameter
      model/llm: choice(OpenAI, GPTTurbo, GPT4)
    # you can also define custom search space for objective function (https://hydra.cc/docs/plugins/optuna_sweeper/#experimental--custom-search-space-optimization)
    # please see ./custom_search_space.py for example
    custom_search_space: configs.hparams_search.custom-search-space-objective.configure