From e443ee312e06c9cb42960915efa3f906c7886eba Mon Sep 17 00:00:00 2001
From: MichalZajacMTS <127204802+MichalZajacMTS@users.noreply.github.com>
Date: Mon, 3 Apr 2023 09:26:38 +0200
Subject: [PATCH] receptive field of view for clustering was not updating
 according to input images size - fixed

---
 python/caffe/layers/detectnet/clustering.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/python/caffe/layers/detectnet/clustering.py b/python/caffe/layers/detectnet/clustering.py
index 9547db7d880..77bec07387b 100644
--- a/python/caffe/layers/detectnet/clustering.py
+++ b/python/caffe/layers/detectnet/clustering.py
@@ -150,7 +150,8 @@ def gridbox_to_boxes(net_cvg, net_boxes, self):
     cell_width = im_sz_x / grid_sz_x
     cell_height = im_sz_y / grid_sz_y
 
-    cvg_val = net_cvg[0, 0:grid_sz_y, 0:grid_sz_x]
+    #cvg_val = net_cvg[0, 0:grid_sz_y, 0:grid_sz_x]
+    cvg_val = net_cvg[0][0:net_cvg.shape[1],0:net_cvg.shape[2]]
 
     if (self.is_groundtruth):
         mask = (cvg_val > 0)