Danil Zherebtsov

Oh, sorry, I mean Warnings that were created by the Profile Report. Please check the screenshot below

Hi, thank you for the answer. Some of the definitions in RuSentiment README are not clear. For example 'neutral class (unmarked for sentiment)' - what does that mean? If this...

Hi Hellorp1990, there had been a few articles on medium about some of `verstack`s modules, here are the links: https://danilzherebtsov.medium.com/parallelise-like-a-boss-with-a-single-line-of-code-in-python-30af0d640511 https://danilzherebtsov.medium.com/automatic-feature-selection-in-python-f72ec69215fe https://danilzherebtsov.medium.com/meet-stacker-automated-ensemble-creation-library-abcad0648a5d https://danilzherebtsov.medium.com/ultimate-python-dateparser-with-holidays-paydays-and-all-the-good-stuff-fffe270a236 https://danilzherebtsov.medium.com/effortlessly-tune-lgbm-with-optuna-49de040d0784 https://danilzherebtsov.medium.com/advanced-categoric-encoding-like-a-pro-onehot-meantarget-woe-frequency-factorization-99dc69477bf3 https://medium.com/swlh/impute-missing-values-the-right-way-c63735fccccd https://danilzherebtsov.medium.com/continuous-data-stratification-c121fc91964b

Is this what you are looking for? And which of the modules have you been using in your research?

@hellorp1990 BTW next generation NaNImputer is getting ready for release. It will be much faster than the current version. If possible, please share a link to you paper when published!

@hellorp1990, the new and upgraded version of `NaNImputer` had been released. It works much faster, has a simpler architecture and is more reliable. You can upgrade to the new version...

> I Geethen. I presume you are referring to the `LGBMTuner` class within `verstack`, right? How big is the dataset you using for training and what is the task type...

For regression tasks every trial of hyperparameters optimisation within LGBMTuner is carried out on a new random split. This way this is very similar to a cross-validation approach you are...

I've been planning to add a holdout-validation option into LGBMTuner for quite some time now, specifically for time series applications. I guess this will be the motivation. I will let...