Skip to content
Edit on GitHub

LightGBM

DVCLive allows you to add experiment tracking capabilities to your LightGBM projects.

Usage

Include the DVCLiveCallback in the callbacks list passed to the lightgbm.train call:

from dvclive.lgbm import DVCLiveCallback

...

lightgbm.train(
  param, train_data, valid_sets=[validation_data], num_round=5,
  callbacks=[DVCLiveCallback()])

Parameters

  • model_file - (None by default) - The name of the file where the model will be saved at the end of each step.

  • live - (None by default) - Optional Live instance. If None, a new instance will be created using **kwargs.

  • **kwargs - Any additional arguments will be used to instantiate a new Live instance. If live is used, the arguments are ignored.

Examples

  • Using live to pass an existing Live instance.
from dvclive import Live
from dvclive.lgbm import DVCLiveCallback

live = Live("custom_dir")

lightgbm.train(
    param,
    train_data,
    valid_sets=[validation_data],
    num_round=5,
    callbacks=[DVCLiveCallback(live=live)])

# Log additional metrics after training
live.summary["additional_metric"] = 1.0
live.make_summary()
  • Using model_file.
lightgbm.train(
    param,
    train_data,
    valid_sets=[validation_data],
    num_round=5,
    callbacks=[DVCLiveCallback(model_file="lgbm_model.txt")])
  • Using **kwargs to customize the new Live instance.
lightgbm.train(
    param,
    train_data,
    valid_sets=[validation_data],
    num_round=5,
    callbacks=[DVCLiveCallback(
      model_file="lgbm_model.txt",
      dir="custom_dir")])
Content

🐛 Found an issue? Let us know! Or fix it:

Edit on GitHub

Have a question? Join our chat, we will help you:

Discord Chat