libcuda | Rust interface to CUDA | GPU library

Instructions to setup Tensorflow 2.4.x (tested for 2.4.1) on an Ubuntu 20.04 environment without admin rights. It is assumed that a sysadmin already installed the latest Cuda drivers. It consists of install Cuda 11.0 toolkit + CuDNN 8.2.0.

Instructions below will install CUDA 11.0 (tested to work for Tensorflow 2.4.1) under directory /home/pherath/cuda_toolkits/cuda-11.0 without sudo rights.

Step 1. Download CUDA 11.0

After reboot everything works ! It seems that after installing cuda & tensorflow I had to restart the PC.

I have had to record a 2D quiver plot a while back.

The approach I used, was:

1. Create the figure you want using.
2. Convert the fig to an image (a numpy array)
3. Use opencv-python (cv2), to write the output.

This could be due to the batch_size in the pipeline.config file. Try reducing it to 1 and see if it works.

You have to update Tensorflow to version 2.3 or greater - that's where the TPUStrategy was added. Here's is your clue to this:

Your problem is created by the linker trying to dynamically load the library. There could be several root-causes for this:

1. Permissions. The user should have permission to load the library, so when mounting file systems in docker, the owner id and the group id that are in the host are not necessary the same id in the container although they might be the same name.
2. Wrong binary format. The host OS is compiling the binary in wrong format. This can happen if you run the compile on (by example) macOS and use it in a linux container.
3. Wrong mounting. The mounting, by example, with noexec will also prevent the library to be loaded.
4. Difference in libraries from both environments. Due to the differences of the environment where the library was compiled, you might be missing some libraries, so use ldd grid_subsampling.cpython-35m-x86_64-linux-gnu.so and ldd -r -d -v grid_subsampling.cpython-35m-x86_64-linux-gnu.so check all the libraries that are linked.

The suggested way to build the most reliable container is to use the officially maintained 'Deep Learning Containers'. I would suggest pulling 'gcr.io/deeplearning-platform-release/tf2-gpu.2-4'. This should already have CUDA, CUDNN, GPU Drivers, and TF 2.4 installed & tested. You'll just need to add your code into it.

• https://cloud.google.com/ai-platform/deep-learning-containers/docs/choosing-container
• https://console.cloud.google.com/gcr/images/deeplearning-platform-release?project=deeplearning-platform-release
• https://cloud.google.com/ai-platform/deep-learning-containers/docs/getting-started-local#create_your_container

As @talonmies pointed out, I was misunderstanding the versioning system. However, because it's a System76 machine, it was also confounding because System76 uses their own Nvidia driver, and it's not straightforward to install Cuda 11 and Cudnn. I'm posting the answer in case anyone else runs into problems with System76.

First, DO NOT use the System76 install for Cuda and Cudnn. They have their own versions (on their website) so as to be compatible with their Nvidia driver, but they will not work (they are version 10, and TF 2.2+ requires 11). Also, most general Cuda guides will tell you to uninstall/install the Nvida driver first so as to have a clean install, but DO NOT do this if you have a System76 system. Just leave the System76 driver alone. Also, if you have any previous Cuda/Cudnn remove/uninstall all of it.

Go to Nvidia and get their latest Cuda and Cudnn. I used

Nothing to worry about. Just burn the system.