[GRASS-SVN] r73170 - grass-addons/grass7/imagery/i.ann.maskrcnn/maskrcnnlib

svn_grass at osgeo.org svn_grass at osgeo.org
Fri Aug 24 08:51:04 PDT 2018


Author: pesekon2
Date: 2018-08-24 08:51:04 -0700 (Fri, 24 Aug 2018)
New Revision: 73170

Modified:
   grass-addons/grass7/imagery/i.ann.maskrcnn/maskrcnnlib/config.py
Log:
new order of shapes settings in config.py

Modified: grass-addons/grass7/imagery/i.ann.maskrcnn/maskrcnnlib/config.py
===================================================================
--- grass-addons/grass7/imagery/i.ann.maskrcnn/maskrcnnlib/config.py	2018-08-24 13:55:57 UTC (rev 73169)
+++ grass-addons/grass7/imagery/i.ann.maskrcnn/maskrcnnlib/config.py	2018-08-24 15:51:04 UTC (rev 73170)
@@ -199,6 +199,10 @@
         # See model.compute_backbone_shapes
         COMPUTE_BACKBONE_SHAPE = None
 
+        # Input image size
+        self.IMAGE_SHAPE = np.array(
+            [self.IMAGE_MAX_DIM, self.IMAGE_MAX_DIM, 3])
+
         # Compute backbone size from input image size
         self.BACKBONE_SHAPES = np.array(
             [[int(math.ceil(self.IMAGE_SHAPE[0] / stride)),
@@ -205,10 +209,6 @@
               int(math.ceil(self.IMAGE_SHAPE[1] / stride))]
              for stride in self.BACKBONE_STRIDES])
 
-        # Input image size
-        self.IMAGE_SHAPE = np.array(
-            [self.IMAGE_MAX_DIM, self.IMAGE_MAX_DIM, 3])
-
         # Train or freeze batch normalization layers
         #  None: Train BN layers in a normal mode
         #  False: Freeze BN layers (recommended for small batch size)



More information about the grass-commit mailing list