/bin/bash: ngc: command not found - new topic (July 2022)

Please provide the following information when requesting support.

• Hardware (GTX 3080)
• Network Type (Classification)
• TLT Version ( dockers:
nvidia/tao/tao-toolkit-tf:
v3.22.05-tf1.15.5-py3:
docker_registry: nvcr.io)
format_version: 2.0
toolkit_version: 3.22.05
published_date: 05/25/2022

• Training spec file(If have, please share here)
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)


This is the issue
I have read the earlier thread and tried your instructions.
As you can see from the screen grab they did not work here.
Has the ngc command been changed recently?

According to
!wget “NVIDIA NGC” -P $LOCAL_PROJECT_DIR/ngccli
!unzip -u “$LOCAL_PROJECT_DIR/ngccli/$CLI” -d $LOCAL_PROJECT_DIR/ngccli/

The ngc should be available in your $LOCAL_PROJECT_DIR/ngccli/

Please check.

Then chmod +x the ngc file. And run it.

I have done that but as this shows:


‘ngc’ is at /home/peter//TAO_toolkit/jp1/ngccli/ngc-cli
it is not at ngc_assests/ngccli/ngc-cli/ngc

Please advise - it doesn’t look like a directory called 'ngc_assets has been created.

In the notebook, there is not such cell. The notebook does not mention “ngc_assets”.
See https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/resources/cv_samples/version/v1.4.0/files/classification/tao_voc/classification.ipynb

Agreed, but the ngc file does not run despite the fact that I have made it executable because ls: cannot access ‘ngc_assests/ngccli/ngc-cli/ngc’: No such file or directory.


Please advise

Can you use “find” command to find where is the “ngc” bin file?

I already uploaded that information in post 2, but here it is again

So, you can use below way.

! /home/peter/TAO_toolkit/jp1/ngccli/ngc-cli/ngc registry model list nvidia/tao/pretrained_classification:*

Thank you, that works. Can you explain the issue?

I am afraid you did not run below well.

Do I need to go back? Does it create other issues?

Never mind. It is not needed.
The ngc is just a tool for you to download pretrained model.
It is not related to TAO training or etc.

Thank you