LIGHTGBM(External algorithm library)¶
LightGBM is a gradient boosting framework that uses tree based learning algorithms.
Running with RecBole¶
convert_token_to_onehot (bool): If True, the token type features will be converted into one-hot form. Defaults to
token_num_threhold (int): The threshold of one-hot conversion. Defaults to
lgb_silent (bool, optional): Whether to print messages during construction. Defaults to
lgb_model (file name of stored lgb model or 'Booster' instance): Lgb model to be loaded before training. Defaults to
lgb_params (dict): Booster params.
lgb_learning_rates (list, callable or None): List of learning rates for each boosting round or a customized function that calculates
learning_ratein terms of current number of round (e.g. yields learning rate decay). Defaults to
lgb_num_boost_round (int): Number of boosting iterations. Defaults to
lgb_early_stopping_rounds (int or None): Activates early stopping. The model will train until the validation score stops improving. Validation score needs to improve at least every
early_stopping_roundsround(s) to continue training. Defaults to
lgb_verbose_eval (bool or int): Requires at least one validation data. If True, the eval metric on the valid set is printed at each boosting stage. If int, the eval metric on the valid set is printed at every
verbose_evalboosting stage. The last boosting stage or the boosting stage found by using
early_stopping_roundsis also printed. Defaults to
Please refer to [LightGBM Python package](https://lightgbm.readthedocs.io/en/latest/Python-API.html) for more details.
A Running Example:
Write the following code to a python file, such as run.py
from recbole.quick_start import run_recbole run_recbole(model='lightgbm', dataset='ml-100k')
If you want to change parameters, dataset or evaluation settings, take a look at