ValueError when Running rsp_rvt

[rcmdnk]

When running rsp_process on my sample, I encountered the following error:

ValueError: Number of samples, -1, must be non-negative.

This error occurs during the RVT calculation, and I discovered that fr_phase has the same values around the corresponding indexes in _rsp_rvt_harrison . By applying the following patch to _rsp_rvt_find_min, the error is resolved:

--- a/neurokit2/rsp/rsp_rvt.py
+++ b/neurokit2/rsp/rsp_rvt.py
@@ -325,7 +325,7 @@ def _rsp_rvt_find_min(increase_inds, fr_phase, smaller_index, silent):
     n_min = increase_inds[bigger_n_max]
     fr_min = fr_phase[n_min].squeeze()
     # Sometime fr_min is the same as n_max and it caused problems
-    if fr_phase[smaller_index].squeeze() < fr_min:
+    if fr_phase[smaller_index].squeeze() <= fr_min:
         if not silent:
             warn(
                 "rsp_rvt(): The next bigger increasing index has a bigger value than the chosen decreasing index, "

But the code contains the following comment:

# Sometime fr_min is the same as n_max and it caused problems

• What specific problems occur when fr_min is the same as n_max?
• Should I avoid this calculation if this condition arises, or can the above patch be applied?

[welcome[bot]]

Hi 👋 Thanks for reaching out and opening your first issue here! We'll try to come back to you as soon as possible. ❤️