Error in levelX[[xIx]]: subscript out of bounds

[nick-youngblut]

model2DE(model = mod, 
                 model_type = 'rf', 
                 data = data[,2:ncol(data)], 
                 target = as.factor(data[,1]), 
                 classPos = 'pos_class', 
                 filter = FALSE, 
                 discretize = FALSE)

Error in levelX[[xIx]]: subscript out of bounds
Traceback:

2. extractDecisions(treeList, X = data, ntree = ntree, maxdepth = maxdepth, 
 .     in_parallel = in_parallel, n_cores = n_cores, cluster = cluster)
3. lapply(1:ntree, extractSingleTree, treeList = treeList, levelX = levelX, 
 .     maxdepth = maxdepth)
4. FUN(X[[i]], ...)
5. treeVisit_endoR(tree, rowIx = rowIx, count, ruleSet, rule, levelX, 
 .     length = 0, max_length = maxdepth)
6. treeVisit_endoR(tree, tree[rowIx, "left daughter"], count, ruleSet, 
 .     ruleleft, levelX, length + 1, max_length)
7. treeVisit_endoR(tree, tree[rowIx, "right daughter"], count, ruleSet, 
 .     ruleright, levelX, length + 1, max_length)
8. treeVisit_endoR(tree, tree[rowIx, "left daughter"], count, ruleSet, 
 .     ruleleft, levelX, length + 1, max_length)
9. treeVisit_endoR(tree, tree[rowIx, "left daughter"], count, ruleSet, 
 .     ruleleft, levelX, length + 1, max_length)
10. treeVisit_endoR(tree, tree[rowIx, "left daughter"], count, ruleSet, 
  .     ruleleft, levelX, length + 1, max_length)

[aruaud]

Hmm I am not sure about this error.

• Can you check if the same dataset was used for fitting the model and endoR? - especially the number and order of columns.
• endoR should convert your data to data.frame automatically, but it may have failed, leading to the error: can you try class(data) <- 'data.frame' before running endoR?
• last thing I can think about: what about using colnames(data) <- compatibleNames(colnames(data)) before endoR, does it solve it?