From f8ec503d6ce5a6ad6b55a7c85ac8817c15f400f8 Mon Sep 17 00:00:00 2001
From: noahnovsak <noah.novsak@gmail.com>
Date: Wed, 16 Aug 2023 12:20:11 +0200
Subject: [PATCH] RemoveNaNColumns computes nans in advance

---
 Orange/preprocess/preprocess.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Orange/preprocess/preprocess.py b/Orange/preprocess/preprocess.py
index 44f7cc6f6f2..6843dc1072c 100644
--- a/Orange/preprocess/preprocess.py
+++ b/Orange/preprocess/preprocess.py
@@ -307,7 +307,8 @@ def __call__(self, data, threshold=None):
                         self.threshold
         if isinstance(threshold, float):
             threshold = threshold * data.X.shape[0]
-        nans = np.sum(np.isnan(data.X), axis=0)
+        # compute nans in advance, otherwise dask will do it for every attribute
+        nans = np.asarray(np.sum(np.isnan(data.X), axis=0))
         att = [a for a, n in zip(data.domain.attributes, nans) if n < threshold]
         domain = Orange.data.Domain(att, data.domain.class_vars,
                                     data.domain.metas)