-
In PyTorch, is there any way to remove the batchnorm layers by merging their parameters into their preceding convolutional layers?
ps: If you are familiar with the model deployment function in MatC…
-
Hello and thak you for your work!
Testing your code for the NYU_v2 trained tensorflow models worked great for me!
But I'd also like to test your model on outdoor scenes. So I installed Matlab and …
-
Does the code (training or testing) compatible with newer version of matconvnet, such as v1.0-beta25?
-
Hi @albanie
Is there any schedule to implement DSSD in matconvnet?
-
Hi
Focal Loss and RetinaNet by the same group as Faster RCNN seems to be an extension of Faster RCNN.
Any plans of implementing it on MatConvNet?
paper:
https://arxiv.org/abs/1708.02002
Best Re…
-
I have Lasagne versions of vgg-16, vgg-19 and googlenet with their pretrained parameters:
http://www.vlfeat.org/matconvnet/pretrained/
https://github.com/BVLC/caffe/tree/master/models/bvlc_googlenet
…
-
Environment: ubuntu16.04 + GTX1660 + cuda8.0 + cudnn6.0 + opencv3.4.1 + matlab2017a.
The version of gcc and g++: 4.9
I have downloaded the code from github.
First step: run the install
the resul…
-
hi, I tried this code but I continue receive this error:
Extracting CNN features for surface normals of depth images...
Warning: Name is nonexistent or not a directory: /home/matt/Desktop/exploitCNN…
-
Hi Justin,
I'm trying to download fast-network-style trained networks models and export them to Matlab to be used by Matconvnet library. I was wondering if that is possible. Also, how to access th…
-
Hi,
My platform is OS X El Capitan, Matlab 2015b, Cuda 7.5 (we won't need this for current issue,though). I can successfully compile the toolbox with CPU support. However vl_test_imreadjpeg.m fails …