Satoshi Nagayasu · @snaga
27 followers · 322 posts · Server mstdn.social

Learning LightGBM for the first time.

#lightgbm #analytics #MachineLearning

Last updated 2 years ago

Markus Lauber · @markks
64 followers · 97 posts · Server det.social

Machine Learning with or gets better with optimisation and a tool like is there to help. Also, you can integrate the results with - a walkthrough with some code (medium.com/p/dcf0efdc8ddf)

#xgboost #lightgbm #hyperparameter #optuna #KNIME

Last updated 2 years ago

Markus Lauber · @markks
64 followers · 114 posts · Server det.social

is a powerful tool. If you combine it with the data science platform and some hyper-parameter tuning it is even better. You can read my new medium blog about how to do that. You can download the code and workflow and try it for yourself. I also employed some along the way - for the fun of it 😀

medium.com/@mlxl/hyperparamete

#lightgbm #machinelearning #LowCode #KNIME #chatgpt

Last updated 3 years ago

Marcin Paprzycki · @marcinpaprzycki
6 followers · 83 posts · Server masto.ai

Presenting results from the method on the forwarding contracts pricing FedCSIS 2022 challenge in: “Using gradient boosting trees to predict the costs of ” by S. Pioroński, T. Górecki. @FedCSIS
2022, ACSIS Vol. 30 p. 421–424; tinyurl.com/bdfwnnn9

#lightgbm #forwardingcontracts

Last updated 3 years ago

mariofilho · @mariofilho
10 followers · 2 posts · Server mastodon.social

You can get to 95% of the performance of a huge ensemble in a tabular dataset by simply running bayesian optimization with 30 steps with a GBM-like model

Set the maximum number of trees you can

Optimize learning rate, max depth, subsample, colsample and minimum samples per leaf

#machinelearning #datascience #kaggle #xgboost #lightgbm

Last updated 3 years ago

Carl Gold, PhD · @carl24k
43 followers · 10 posts · Server sigmoid.social

lesson of the day: Working with a model, I got no traction cross-validating hyperparameters like tree depth and number; but different evaluation metrics (e.g. SMSE vs. MAE etc.) had a major impact. Have you tried this?

IMHO get all the press because they do sexy human jobs like seeing and processing language. But in the business world of tabular data, is where the real revolution is happening!

#machinelearning #gradientboosting #dnn #xgboost #catboost #lightgbm #DataScience

Last updated 3 years ago

Lubomír Šerý · @gugatr0n1c
8 followers · 7 posts · Server mastodonczech.cz