Python 3.8 and tensorflow 2.10 - DeepTalk - Deep Learning Community
Python 3.8 and tensorflow 2.10
post by cmachreki on Aug 30, 2023
cmachreki
is there a way for me to install both of those and use a gpu say the T4 on a vm , all to train ML models? please advise and help!
post by markd on Sep 5, 2023
markd
Was this meant to be part of a thread? And is this your own machine you are running a VM on? The cloud is using VMs.
So yes you will need Python and tensorflow installed for tensorflow based project.
You can ‘version’ using Docker, Python venv or Anaconda/miniconda. And that can be done on baremetal machines, VMs or cloud.
We have simple ‘getting started’ documentation for these at:
- Linux | Lambda Docs
- How do I install Docker and run a container? | Lambda Docs
- How do I create a Python virtual environment? | Lambda Docs
- How do I create a conda virtual environment? | Lambda Docs
A VM adds overhead, and docker would be the simplest “pre-canned” images.
NVIDIA provides the NGC repository which could be helpful to pick python version + TF version.
Next easiest is Anaconda/Conda to use for a old version of Python no longer installed.
conda create --name my_env python=3.8.5 -c anaconda -c nvidia
python venv the old version should be a alternate Ubuntu version installed.
post by cmachreki on Sep 5, 2023
cmachreki
Dear Mark,
Thanks for your reply, although a bit late but thank you.
God Bless Brother.
Related topics
| Topic | Replies | Views | Activity |
|---|---|---|---|
| Tensorflow in virtual environment on GPU Workstation | 1 | 1.7k | Nov 2020 |
| Updating Stack on GPU Cloud to latest versions | 1 | 1.2k | Apr 2023 |
| Jupyter notebook having old version of tensorflow | 0 | 1.3k | May 2020 |
| Tensorflow 1.13 and theano 1.0.4 not using gpu | 4 | 3.3k | Jul 2019 |
| Tensorflow does not detect GPU | 1 | 3.8k | May 2021 |