Erro while pulling the TLT

I’m getting this after succesfully login to nvcr.io:

Error response from daemon: pull access denied for nvcr.io/nvtltea/med/tlt-annotate, repository does not exist or may require ‘docker login’

Any ideas what might be wrong?

P.S: I was not never asked to select an organization while creating my NCG account.

The transfer learning toolkit is part of Clara Train and requires a formal approval process. Did you apply for the early access (here: https://developer.nvidia.com/clara)?

Hi Nrieke,

Does TLT still need to apply for early access? I think Clara Train/Deploy SDK has opened to access, but when I follow this article:

and try to run this command:

docker run --runtime=nvidia -it -p 5000:5000 -v /path/to/save/model:/var/tmp/models -e NVIDIA_VISIBLE_DEVICES=0 nvcr.io/nvtltea/med/tlt-annotate:v0.1-py3 /bin/bash

It still shows:

docker: Error response from daemon: pull access denied for nvcr.io/nvtltea/med/tlt-annotate, repository does not exist or may require ‘docker login’: denied: requested access to the resource is denied.

Does the TLT still need to apply for the access? If so, I have submitted for approval some days ago, how long it will take to approve.
Or is there some other way to download the trained model or this library.

Thank you very much.

Hi
Thanks for your interest in clara train sdk. the SDK is now open since the GA release v1.0. We also just released v1.1 that you can download WITHOUT registration or even login from ngc
https://ngc.nvidia.com/catalog/containers/nvidia:clara-train-sdk

Also please see our updated documentation showing new feature as bring your own components (loss, model architecture, reader, transformation, metric) at
https://docs.nvidia.com/clara/tlt-mi/clara-train-sdk-v1.1/index.html