AllardJM

Results 13 comments of AllardJM

Upgrading to 1.0.0-rc2 results in a large drop in classification performance using LightGBMClassifier.

@imatiach-msft Can you point me to the latest jar file that fixes this issue?

from __future__ imports must occur at the beginning of the file

Can you explain this fix a bit more @simonljus ?

from __future__ imports must occur at the beginning of the file

Thanks! I will review this and see if it works for me.

Large prediction results unless using repartition(1) in databricks with lgbm model

I am seeing something similar - enormous predictions that are sometimes positive and sometimes negative (there are no negative values in the target). It seems that if I use mse...

[BUG]An error occurred while calling o1156.fit. : java.lang.UnsatisfiedLinkError:

It would be great if there is some advice on this - how can EMR run the newer versions of LGBM (I am interested in the streaming execution)?

[LightGBM] Classifier hangs on fit most of the time

I am using 11.2 and experiencing similar issues. Fit hangs before starting. It will run on really small datasets but fail on anything larger.

Different prior distributions for different media channels on same coefficient

@l-d96 Did you ever find a way to do this? Even at the basic level, how do you pass in different parameters for the same distribution for each of the...

Optimizer Fails to Find Best Values

Looking into the source code, it seems that part of this issue is related to the following: From the objective function: ``` spend = np.full(self.num_periods, budget) spend_extended = np.concatenate([spend, np.zeros(self.adstock.l_max)])...

plot_direct_contribution_curves doc string incorrectly states the plot does not include lagged effects

@timbo112711 @cluhmann Can I restate the consensus on what this plot actually shows? I am finding it hard to tell on numerous artifacts within the package and this is one....

SNOW-1648322: ML model registration doesn't work when the package name has dot in it

I'm experiencing this same issue trying to register a Pytorch model. The exact same notebook ran last week - using the GPU runtime that is exposed with SF Notebooks... SnowparkSQLException:...