[Tune] Feature request: Using Distribute Ray Tune for K-Fold on lightgbm

Scott_Zhang · August 26, 2024, 3:53am

I was trying to use Ray Tune for K-Fold with distribute mode on lightgbm.

As i read lightgbm_example file, it only support local tune with origin lightgbm api.
with large sample data, it will make oom.

github.com

ray-project/ray/blob/master/python/ray/tune/examples/lightgbm_example.py

import lightgbm as lgb
import sklearn.datasets
import sklearn.metrics
from sklearn.model_selection import train_test_split

from ray import tune
from ray.tune.integration.lightgbm import TuneReportCheckpointCallback
from ray.tune.schedulers import ASHAScheduler


def train_breast_cancer(config: dict):
    # This is a simple training function to be passed into Tune

    # Load dataset
    data, target = sklearn.datasets.load_breast_cancer(return_X_y=True)

    # Split into train and test set
    train_x, test_x, train_y, test_y = train_test_split(data, target, test_size=0.25)

    # Build input Datasets for LightGBM

This file has been truncated. show original

Is it possible to include a new tune API, support distribute cv tune on lightgbm?

Topic		Replies	Views
Tune Performance issue with LightGBM predict Ray Tune	2	1138	November 10, 2022
Tune Performance Issue with LightGBM predict Ray Tune	0	269	November 7, 2022
[Tune] Feature request: Using Ray Tune for K-Fold with Repeated Grid Sampling Ray Tune	3	601	June 17, 2021
Where do I find documentation on the tune.run method Ray Libraries (Data, Train, Tune, Serve)	3	2037	June 12, 2023
Tune xgboost with cross-validation? Ray Tune	2	763	September 22, 2021

[Tune] Feature request: Using Distribute Ray Tune for K-Fold on lightgbm

Related topics