This is the suggested way to install the dependencies on Ubuntu 14.04+:
- NVIDIA CUDA 7.0+
- NVIDIA cuDNN v4
- Torch
- NVIDIA GPU with compute capability 3.5 or above
- Install the
build-essential
package:
sudo apt-get install build-essential
- If you are using a Virtual Machine (like Amazon EC2 instances), install:
sudo apt-get update
sudo apt-get install linux-generic
-
Download the CUDA .deb file for Linux Ubuntu 14.04 64-bit from: https://developer.nvidia.com/cuda-downloads. The file will be named something like
cuda-repo-ubuntu1404-7-5-local_7.5-18_amd64.deb
-
Install CUDA from the .deb file:
sudo dpkg -i cuda-repo-ubuntu1404-7-5-local_7.5-18_amd64.deb
sudo apt-get update
sudo apt-get install cuda
echo "export PATH=/usr/local/cuda/bin/:\$PATH; export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:\$LD_LIBRARY_PATH; " >>~/.bashrc && source ~/.bashrc
- Restart your computer
-
Download cuDNN v4 from https://developer.nvidia.com/cuDNN (requires registration). The file will be named something like
cudnn-7.0-linux-x64-v4.0-rc.tgz
. -
Extract the file to
/usr/local/cuda
:
tar -xvf cudnn-7.0-linux-x64-v4.0-rc.tgz
sudo cp cuda/include/*.h /usr/local/cuda/include
sudo cp cuda/lib64/*.so* /usr/local/cuda/lib64
- Install the Torch dependencies:
curl -sk https://raw.githubusercontent.com/torch/ezinstall/master/install-deps | bash -e
- Install Torch in a local folder:
git clone https://github.com/torch/distro.git ~/torch --recursive
cd ~/torch; ./install.sh
If you want to uninstall torch, you can use the command: rm -rf ~/torch
git clone -b R4 https://github.com/soumith/cudnn.torch.git
cd cudnn.torch; luarocks make
sudo apt-get update
sudo apt-get install -y libhdf5-serial-dev hdf5-tools
git clone https://github.com/davek44/torch-hdf5.git
cd torch-hdf5
luarocks make
The RGB and depth images have been properly preprocessed (including synchronization and projection of depths to RGB image planes) for both NYU-Depth-V2 and KITTI datasets. They are stored in HDF5 formats, with each file containing one RGB-depth image pair. This might take an hour or so.
cd data
wget http://datasets.lids.mit.edu/sparse-to-dense/data/kitti.tar.gz
tar -xvf kitti.tar.gz && rm -f kitti.tar.gz
wget http://datasets.lids.mit.edu/sparse-to-dense/data/nyudepthv2.tar.gz
tar -xvf nyudepthv2.tar.gz && rm -f nyudepthv2.tar.gz
cd ..
git clone https://github.com/fangchangma/sparse-to-dense.git
cd sparse-to-dense