Fix issue with VGG16 from slim. The slim models adds the hot-encoded in the architecture function
Compare changes
+ 63
− 2
@@ -9,7 +9,10 @@ VGG16 and VGG19 wrappers
@@ -9,7 +9,10 @@ VGG16 and VGG19 wrappers
@@ -40,7 +43,9 @@ def vgg_19(inputs,
@@ -40,7 +43,9 @@ def vgg_19(inputs,
@@ -57,9 +62,65 @@ def vgg_16(inputs,
@@ -57,9 +62,65 @@ def vgg_16(inputs,