An open API service indexing awesome lists of open source software.

https://github.com/danindiana/tensorcatgpu

Can CPU tensorflow build CUDNN tensorflow?
https://github.com/danindiana/tensorcatgpu

bash-scripting cudnn-based-implementations nvidia-gpu tensorflow

Last synced: 3 months ago
JSON representation

Can CPU tensorflow build CUDNN tensorflow?

Awesome Lists containing this project

README

        

# tensorcatgpu
Nvidia driver + depdenencies GPU compiled tensorflow (NDGPUTF) on Ubuntu is really challenging. Is this difficulty itself a kind of intelligence test?

What are Neural Networks good for anyway?

Can tensorflow CPU (CPUTF) be used to assist in installing / compiling Nvidia GPU (CUDNN) supported tensorflow?

Can you use tensorflow CPU to bootstrap a fault-tolerant Nvidia GPU (CUDNN) supported tensorflow build?
Can you use tensorflow CPU to bootstrap a fault-tolerant AMD GPU (RocM) supported tensorflow build?

Assume you are trying to install on an Ubuntu 18 OS with following system specs:

OS: Ubuntu 18.10 cosmic
Kernel: x86_64 Linux 4.18.0-15-generic
Uptime: 6h 43m
Packages: 2412
Shell: bash 4.4.19
Resolution: 2560x1080
DE: LXDE
WM: OpenBox
CPU: AMD Ryzen 5 1600X Six-Core @ 12x 4GHz [36.6°C]
GPU: GeForce GTX 1050
RAM: 8732MiB / 16034MiB

Yes, you do need a developer account with nvidia to get the necessary files like: libcudnn7-dev_7.4.2.24-1+cuda10.0_amd64.deb

To DO: see further readings from cat tf_env.txt file.

Install TensorFlow CPU
https://www.tensorflow.org/install/pip

Describe the neural network or how to generate such for the Ubuntu Linux system.

GPT output:
The neural network that would need to be generated for the Ubuntu Linux system would be a Convolutional Neural Network (CNN). The input to the CNN would be information about the OS, kernel, uptime, packages, shell, resolution, DE, WM, CPU, GPU and RAM of the system. The output of the network would be installation commands for setting up a bare metal GPU supported TensorFlow instance.