Mozhdeh Dokhani

> There are too many hyper-parameters on train.lua and it is hard to train the model as good as the pretrained checkpoint model. > > The model I trained only...

It was due to the gpu memory size. I run it on some othe computer with higher gpu memory size. Close the issue

Today I added this line to .bashrc file: ///////////////////////////////////////////////////////////////// export CUDNN_PATH=/usr/local/cuda-9.1/lib64/libcudnn.so.7 ///////////////////////////////////////////////////////////////// and the error changed to: ///////////////////////////////////////////////////////////////// Found Environment variable CUDNN_PATH = /usr/local/cuda-9.1/lib64/libcudnn.so.7/home/mozhdeh/torch/install/bin/luajit: /home/mozhdeh/torch/install/share/lua/5.1/trepl/init.lua:389: /home/mozhdeh/torch/install/share/lua/5.1/trepl/init.lua:389: /home/mozhdeh/torch/install/share/lua/5.1/cudnn/ffi.lua:1743: /usr/local/cuda-9.1/lib64/libcudnn.so.7: cannot open...

Yeah, user the command below: git clone https://github.com/soumith/cudnn.torch.git -b R7 && cd cudnn.torch && luarocks make cudnn-scm-1.rockspec and it'll be fixed :)

Hi, I faced the same problem. I've downloaded MS COCO 2014 train and val. I have also downloaded the VGG_ILSVRC_16_layers.caffemodel. But when I run the command below, the out of...