Michael

Hi @WillianFuks, thanks for the quick and thorough reply! Do you have any idea why R is faster? I suspect the R and TF implementation are both written in a...

I had the same issue. Setting network = gce_make_network("mynetwork", subnetwork = "mysubnet") didn't resolve the issue at first. This is because per default gce_vm(...) has open_webports = T. This leads...

I mentioned [here](https://github.com/amdelamar/jhash/issues/24#issuecomment-742670235l) that for me, it works on Android. I'm using jhash 2.2.0

For me it works great on Android. Thanks @amdelamar for this great lib. To make it work, I have set minSdkVersion 26 and added the following to my build.gradle (app):...

The models get appended to the `models` list. check this out: ``` m = Matcher(test, control, yvar="population", exclude=['label']) m.fit_scores(balance=True, nmodels=10) ``` Now you can access all 10 models. To return...

I don't think I understand you. Can you please be more specific? With the method I mentioned you can in fact look at all information regarding the GLM. That includes...

I think this might be the error you get when there is no categorical data in your predictors. Did you check that you have any data of type `categorical` in...

@JohannesWiesner check out the last commit in my fork. It should fix the problem: The issue is in constructing the formula for `patsy` which doesn't allow for white spaces in...

Maybe, the true prize are the GitHub repos we've created along the way... :D

> A use case here is we're deploying to AWS lambda using their container feature. Installing things outside the system folder is a pain... > > I'm thinking `sync` is...