remove penalty metrics

[ayeletis]

remove penalty metrics

[ayeletis]

One lingering error:

I get this:
Error in tune_balanced_causal_forest(X, Y, W, Y.hat, W.hat, sample.weights = sample.weights, :
unused argument (target.weight.penalty.metric = "split_l2_norm_rate")

[izahn]

I can't seem to reproduce it, what triggered that error?

[ayeletis]

If I run the balanced_causal_forest providing any input for "tune.paramters" I get this error. for example, if I add the last row to the sample code:

fit_causal_beat <- balanced_causal_forest(X_train, Y_train, W_train,
                                     target.weights = as.matrix(Z_train),
                                     target.weight.penalty = my_penalty,
                                     num.trees = num_trees,
                                     tune.parameters=c("min.node.size"))

[izahn]

I believe b810227 should do it, but I'm leaving this issue open for now in case we find any other problems.

[ayeletis]

i removed the packaged, and reinstalled, but I still get the same error:
Error in tune_balanced_causal_forest(X, Y, W, Y.hat, W.hat, sample.weights = sample.weights, :
unused argument (target.weight.penalty.metric = "split_l2_norm_rate")

Thanks!

[izahn]

Hmm, I really cannot reproduce it now. Can you please

1. remove.packages("beat")
2. restart R
3. install beat
4. run the sample usage code with the tune.parameters modification above
   ?

If it gives an error (it does not for me) please run
traceback()
and paste the result here.

[ayeletis]

restarting R was key! it is indeed resolved. we will continue our QA for a bit and let you know if any other issues come up. I'll leave this one open in the meantime in case there are other problems.