-
# CuDNN v5.1 speeds up 3x3 convolution substantially.
Now I'm using MATLAB R2015b with MatConvNet beta19. I was able to compile it but saw errors at runtime:
```
Error using vl_nnconv
forward: cuDN…
-
hi guys, I am a newer in python and caffe, I want to transfer a pre-trained caffemodel to matconvnet format, could anyone tell me how to use import-caffe.py? what are the inputs? Thanks a lot.
ghost updated
6 years ago
-
thanks Joao for this awesome repository :)
Autonn saves all variables in network evaluation, including those in layers which do not need the forward pass for derivative computation, like reshape, r…
-
Hello! I tested **SiamFC-3s-color-pretrained** model on OTB2013 with your python-ported eval code. However, the OPE AUC is about 0.5868, lower than reported 0.608 in SiamFC paper, And the tracker MEE…
-
beta23 released a fast-rcnn demo and corresponding model, which is a milestone, i.e. matconvnet begins to support object detection domain. Is there any plan to further support Faster RCNN in the futur…
-
Hi,
I was running the importer.sh code and I am getting this weird error. Could u please help me so that i can mitigate this problem.
Thanks
I am using python 3.5.2 interpreter in my ubuntu mach…
-
```
% adapt for different input sizes (we end up using 500x500)
if all(opts.inputSize==500)
ctblk = dagnn.ConvTranspose('size', size(filter), 'upsample', 2, ...
…
-
Hello I am trying to train hr_res101('train') with Wider face data. I compiled a CPU only version with matlab 2016b and matconvnet in ubuntu 14.04. After running the hr_res101('train') script I am get…
-
Hi all,
I am trying to re-implement my old matcovnet classification work using mxnet. The initial learning rate is set to 1e-6 in matconvnet, however when I use the same small learning rate in mxnet,…
-
Hello,@aimerykong ,you mentioned "**The folder named "caffe-20160312" is a modified caffe toolbox by Yang Gao for his "compact bilinear pooling**"" in readme.md. I want to know where are you modified,…