You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Add an example page, like cross validation, to document how to perform group-wise tests on non-grouped rotations, like SVD, as well as say in a sentence why these tests are not automatically run (because the rotation does not know what to group by)
import HypothesisTests: KruskalWallisTest, VarianceEqualityTest
import EpistemicNetworkAnalysis: GroupwiseCoregistrationTest, test!
functionaddGroupedTests!(model, groupVar)
# Parameters used by EpistemicNetworkAnalysis.jl to determine which version of test! to run
M =typeof(model)
trainmodel = model
testmodel = model
tests = [GroupwiseCoregistrationTest, KruskalWallisTest, VarianceEqualityTest]
# Find group labels
groups =sort(unique(trainmodel.metadata[!, groupVar]))
# Run tests on each axisfor i in1:nrow(testmodel.embedding)
for test in tests
test!(M, trainmodel, testmodel, test, dim=i, groupVar=groupVar, groups=groups)
endendend
The text was updated successfully, but these errors were encountered:
Add an example page, like cross validation, to document how to perform group-wise tests on non-grouped rotations, like SVD, as well as say in a sentence why these tests are not automatically run (because the rotation does not know what to group by)
This can be based on how it's done in AbstractGroupDifference already
The text was updated successfully, but these errors were encountered: