More things to resolve suggested by Stan

smartdata-analysis-and-statistics / precmed

A doubly robust precision medicine approach to estimate and validate conditional average treatment effects

Apache License 2.0

4 stars 0 forks source link

Closed jcaldasmagalhaes closed 1 year ago

jcaldasmagalhaes commented 1 year ago

script	line	Error	Resolve
outputs_count.R	line 263	# TODO: now score.method has no default (mandatory argument)	Set to “poisson”?
CATE_count.R	253 & 333	# TODO: if model has a single predictor, GBM must have cv.folds = 0 https://github.com/zoonproject/zoon/issues/130
outputs_continuous.R	287	##TODO: Insert n.trees.rf
	570	# TODO: is it right?
	1032	# TODO: Phoebe I think we removed the arg.check on interactions but here it is?
	1043	#TODO: Needs to be specified
	1088	#TODO: modify this..can we use chisq?
Plots.R	635	# TODO: Temporary solutions as cbind not working when cv.n = 1

NightlordTW commented 1 year ago

feedback gabrielle:

outputs_count.R => score.method does not need a default
CATE_count => we need to add a test for this, throw a warning/error for this situation
outputs_continuous => We can ignore for now
Plots.R => force vector as a matrix so we can do cbind?

StanWijn commented 1 year ago

We fixed all 4 points:

Thomas already removed the todo from the output_count.R (score.method)
The error in CATE_count is now fixed, but will result in warnings (as described in #35 )
Plots.R -> tried to force the vector as a matrix, but this will undo the transpose and result in a vertical matrix (i.e., cbind will no longer result in the desired horizontal vector). This could be fixed by transposing the matrix again, but when we run into problems when cv > 1. When looking in to this; there are a couple of work-arounds, but these are complex or require additional packages. I suggest we leave it as is.