Skip to content

Instantly share code, notes, and snippets.

Shruti Mittal MittalShruti

View GitHub Profile
@MittalShruti
MittalShruti / libcudnn.sh
Last active May 10, 2020
libcudnn.so error
View libcudnn.sh
# move to the lib64 directory
cd /usr/local
cd cuda
cd lib64
# check if there are any symlinks
ls -lha libcudnn*
sudo rm libcudnn.so
sudo rm libcudnn.so.7
View cudnn.sh
#Fetch cuDNN
CUDNN_TAR_FILE="cudnn-10.0-linux-x64-v7.6.5.32.tgz"
wget https://developer.nvidia.com/compute/machine-learning/cudnn/secure/7.6.5.32/Production/10.0_20191031/cudnn-10.0-linux-x64-v7.6.5.32.tgz
sudo tar -xzvf ${CUDNN_TAR_FILE}
#copy the files in cuda toolkit directory
sudo cp cuda/include/cudnn.h /usr/local/cuda/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*
View cuda_install.sh
#Get PPA repo drivers and Install CUDA 10.0
curl -O http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_10.0.130-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1604_10.0.130-1_amd64.deb
sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda-10-0 cuda-drivers -y
#Set the environment for cuda 10.0
nano ~/.bashrc
export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}$
@MittalShruti
MittalShruti / packages.sh
Last active May 9, 2020
Install python packages
View packages.sh
#install pip ipython
sudo apt-get -y install python3-pip
sudo pip3 install --upgrade pip
sudo apt-get -y install ipython
#install numpy matplotlob
sudo pip3 install numpy scipy matplotlib ipython jupyter pandas sympy nose
#install tensorflow
sudo pip install tensorflow-gpu
@MittalShruti
MittalShruti / install.sh
Created May 9, 2020
Install CUDA 10.0 cuDNN 7.6.5 Ubuntu 16.04
View install.sh
#Get PPA repo drivers and Install CUDA 10.0
curl -O http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_10.0.130-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1604_10.0.130-1_amd64.deb
sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda-10-0 cuda-drivers -y
#Install cuda-compatibility 10.0
sudo nvidia-smi -pm 1
sudo apt-get install cuda-compat-10.0
@MittalShruti
MittalShruti / install.py
Created May 6, 2020
Install pycodec2 codec2
View install.py
!apt-get update && \
apt-get upgrade -y && \
apt-get install -y git make cmake python3-pip
!git clone https://github.com/drowe67/codec2.git
!cd codec2 && \
mkdir build_linux && \
cd build_linux && \
cmake .. && \
make && \
You can’t perform that action at this time.