Running Transfer Learning Toolkit on local machine

Accelerated Computing Intelligent Video Analytics TAO Toolkit
#1

Hi Guys,

I would like to know if it is possible to run Transfer Learning Toolkit on local machine or can it be only run on cloud?

Thanks.

#2

TLT can run at local machine via docker. See more details in released doc.Thanks.

#3

Can we run TLT on Jetson Nano with JetPack?

#4

Hi rsiu,
See https://docs.nvidia.com/metropolis/TLT/tlt-getting-started-guide/index.html#requirements , using the Transfer Learning Toolkit should meet the hardware and software requirements.

For deployment etlt model or TRT engine in DS at Jetson Nano, please see https://devtalk.nvidia.com/default/topic/1065558/transfer-learning-toolkit/trt-engine-deployment/

#5

Hi Morganh,

Thank you for your reply. I take it that we cannot run TLT on Jetson Nano then. Is there any tutorial on running TLT on AWS or other cloud provider?

BR,
Ricardo

#6

Hi rsiu,
See tlt doc, especially https://docs.nvidia.com/metropolis/TLT/tlt-getting-started-guide/index.html#requirements, you can download the docker and run it in your Ubuntu PC.

#7

Hi Morganh,

I do not have a qualified local machine. I want to run this on the cloud (e.g. AWS). Any pointer?

BR,
Ricardo

#8

You can try if you can run in ngc cloud (ngc.nvidia.com).

#9

How do you do that? Any tutorial? This is my NGC account and what I see after login:

https://drive.google.com/file/d/10fMHMmaciHy_uI5BHJ6h3uJOE6l7s5ON/view?usp=sharing

#10

Below is a good start point for ngc.
https://docs.nvidia.com/deeplearning/frameworks/user-guide/index.html

#11