Merge pull request #3635 from alejoe91/fix-hybrid-example

Fix precision/recall in hybrid example

doc/how_to/benchmark_with_hybrid_recordings.rst

@@ -2531,9 +2531,8 @@ Although non of the sorters find all units perfectly, ``Kilosort2.5``,
 ``Kilosort4``, and ``SpyKING CIRCUS 2`` all find around 10-12 hybrid
 units with accuracy greater than 80%. ``Kilosort4`` has a better overall
 curve, being able to find almost all units with an accuracy above 50%.
-``Kilosort2.5`` performs well when looking at precision (finding all
-spikes in a hybrid unit), at the cost of lower recall (finding spikes
-when it shouldn’t).
+``Kilosort2.5`` performs well when looking at precision (not finding spikes
+when it shouldn’t), but it has a lower recall (finding all spikes in the ground truth).
 
 In this example, we showed how to:
 

examples/how_to/benchmark_with_hybrid_recordings.py

@@ -276,7 +276,8 @@
 # From the performance plots, we can see that there is no clear "winner", but `Kilosort3` definitely performs worse than the other options.
 #
 # Although non of the sorters find all units perfectly, `Kilosort2.5`, `Kilosort4`, and `SpyKING CIRCUS 2` all find around 10-12 hybrid units with accuracy greater than 80%.
-# `Kilosort4` has a better overall curve, being able to find almost all units with an accuracy above 50%. `Kilosort2.5` performs well when looking at precision (finding all spikes in a hybrid unit), at the cost of lower recall (finding spikes when it shouldn't).
+# `Kilosort4` has a better overall curve, being able to find almost all units with an accuracy above 50%. `Kilosort2.5` performs well when looking at precision (not finding spikes
+# when it shouldn’t), but it has a lower recall (finding all spikes in the ground truth).
 #
 #
 # In this example, we showed how to: