fix off-by-one error in euclidean distance computation

[mirka-henninger]

This fixes #162. The computation of the non-gower (e.g., euclidean) distance function was off-by-one as the dist() function already only gives the off-diagonal components of the distance matrix.

[christophM]

Thanks Mirka, good catch.

If you have time, could you maybe add a test for this behavior? You could use the example from #162. If you don't have the time, I would just merge the pull request and add one later.

[mirka-henninger]

Hi Christoph,
Thanks for taking a look at that.
I added a test for the euclidean function that was used for fitting the LocalModel and for all distance functions using get.weight.fun().
Best, Mirka

[codecov-io]

Codecov Report

Merging #163 (fff827b) into master (f1edf3b) will increase coverage by 0.11%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #163      +/-   ##
==========================================
+ Coverage   90.88%   91.00%   +0.11%     
==========================================
  Files          17       17              
  Lines        1667     1667              
==========================================
+ Hits         1515     1517       +2     
+ Misses        152      150       -2     

Impacted Files	Coverage Δ
R/LocalModel.R	97.59% <100.00%> (+2.40%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f1edf3b...fff827b. Read the comment docs.

[christophM]

Perfect, thanks!

[pat-s]

@mirka-henninger Would you mind cleaning up this file with respect to coding style?
There are superfluous whitespace introduced here and some indentions are totally off.

You can make your life easier by calling styler::style_pkg() or styler::style_file() after you're done.

Not sure if more files were touched - appreciated if you could include those as well if that's the case :)