Text Summarization using Transformer on GPU Docker Deployment
Deploying a Text Summarization NLP use case on Docker Container Utilizing Nvidia GPU, to setup the enviroment on linux machine follow up the below process, make sure you should have a good configuration system, my system specs are listed below(I am utilizing DataCrunch Servers) :
- GPU : 2xV100.10V
- Image : Ubuntu 20.04 + CUDA 11.1
Some Insights/Explorations
If you’re a proper linux user make sure to setup it CUDA, cudaNN and Cuda Toolkit
If you’re a WSL2 user then you will face a lot of difficulty in accelarating GPU of host system on WSL, as it has some unknown bugs which are needed to be fixed by them.
After setting up the CUDA and cudaNN,