The FILTERS depth does not divide the DATA depth.
Issue #26
new
I have trained a bcnn-model.But when I run the bird_demo to test it ,this error appeared.
0.05s to load imdb.
i =
195
1.29s to load models into memory. Top 5 prediction for test_image.jpg: 161.Blue_winged_Warbler 178.Swainson_Warbler 025.Pelagic_Cormorant 141.Artic_Tern 137.Cliff_Swallow 0.18s to make predictions [GPU=1]
i =
196
1.43s to load models into memory. Error using vl_nnconv The FILTERS depth does not divide the DATA depth.
Error in vl_bilinearnn (line 214) res(i+1).x = vl_nnconv(res(i).x, l.weights{1}, l.weights{2}, ...
Error in get_bcnn_features (line 95) res = vl_bilinearnn(net, im_resizedA, [], res, ...
Error in bird_demo (line 54) code = get_bcnn_features(net, im, ...