Unable to use tlt pretrained model on NGC

[eybae]

Hi,
I’m using TLT server on tlt(version 3.0) docker to train a detection model.
But the problem is,
When I ran following command in

ngc registry model list nvidia/tlt_pretrained_object_detection:*

I got the result :

±------±------±------±------±------±------±-----±------±------+
| Versi | Accur | Epoch | Batch | GPU | Memor | File | Statu | Creat |
| on | acy | s | Size | Model | y Foo | Size | s | ed |
| | | | | | tprin | | | Date |
| | | | | | t | | | |
±------±------±------±------±------±------±-----±------±------+
±------±------±------±------±------±------±-----±------±------+

I can’t get any pretrained model.
Logining to NGC is successfully so that I can see tao-pretrained-model like below.
ngc registry model list nvidia/tao/pretrained_detectnet_v2:*
±------±------±------±------±------±------±------±------±------+
| Versi | Accur | Epoch | Batch | GPU | Memor | File | Statu | Creat |
| on | acy | s | Size | Model | y Foo | Size | s | ed |
| | | | | | tprin | | | Date |
| | | | | | t | | | |
±------±------±------±------±------±------±------±------±------+
| vgg19 | 82.6 | 80 | 1 | V100 | 153.8 | 153.7 | UPLOA | Aug |
| | | | | | | 7 MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| vgg16 | 82.2 | 80 | 1 | V100 | 113.2 | 113.2 | UPLOA | Aug |
| | | | | | | MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| squee | 65.67 | 80 | 1 | V100 | 6.5 | 6.46 | UPLOA | Aug |
| zenet | | | | | | MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| resne | 82.7 | 80 | 1 | V100 | 294.5 | 294.5 | UPLOA | Aug |
| t50 | | | | | | 3 MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| resne | 77.04 | 80 | 1 | V100 | 170.6 | 170.6 | UPLOA | Jan |
| t34 | | | | | 5MB | 5 MB | D_COM | 28, |
| | | | | | | | PLETE | 2022 |
| resne | 79.0 | 80 | 1 | V100 | 89.0 | 89.02 | UPLOA | Aug |
| t18 | | | | | | MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| resne | 79.2 | 80 | 1 | V100 | 38.3 | 38.34 | UPLOA | Aug |
| t10 | | | | | | MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| mobil | 77.5 | 80 | 1 | V100 | 5.1 | 5.1 | UPLOA | Aug |
| enet_ | | | | | | MB | D_COM | 24, |
| v2 | | | | | | | PLETE | 2021 |
| mobil | 79.5 | 80 | 1 | V100 | 13.4 | 13.37 | UPLOA | Aug |
| enet_ | | | | | | MB | D_COM | 24, |
| v1 | | | | | | | PLETE | 2021 |
| googl | 82.2 | 80 | 1 | V100 | 47.7 | 47.74 | UPLOA | Aug |
| enet | | | | | | MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| effic | 77.11 | 80 | 1 | V100 | 16.9 | 16.9 | UPLOA | Aug |
| ientn | | | | | | MB | D_COM | 18, |
| et_b0 | | | | | | | PLETE | 2021 |
| _swis | | | | | | | | |
| h | | | | | | | | |
| effic | 77.11 | 80 | 1 | V100 | 16.9 | 16.9 | UPLOA | Aug |
| ientn | | | | | | MB | D_COM | 18, |
| et_b0 | | | | | | | PLETE | 2021 |
| _relu | | | | | | | | |
| darkn | 76.44 | 80 | 1 | V100 | 467.3 | 467.3 | UPLOA | Aug |
| et53 | | | | | | 2 MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
| darkn | 77.52 | 80 | 1 | V100 | 229.1 | 229.1 | UPLOA | Aug |
| et19 | | | | | | 5 MB | D_COM | 24, |
| | | | | | | | PLETE | 2021 |
±------±------±------±------±------±------±------±------±------+

Is nvidia-tlt package deprecated completely ?
I don’t know why tlt pretrained model is not appeared.
Can I only use tao package for pretrained model?

Thanks.

[Morganh]

You can use below command.
https://docs.nvidia.com/tao/tao-toolkit/text/tao_toolkit_quick_start_guide.html#listing-all-available-models

ngc registry model list nvidia/tao/pretrained_*

[eybae]

I mean not using TAO.
Is there any way only for TLT?
The command ‘ngc registry model list nvidia/tlt_pretrained_object_detection:*’ workd normally before.

[Morganh]

There is no update from you for a period, assuming this is not an issue anymore.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

TAO is just the renaming of tlt.
For the pretrained model, you can use the latest command to download.