How to train mmj_genai model with our custom dataset of Nvidia metropolis services?

Accelerated Computing Intelligent Video Analytics Metropolis Microservices for Jetson
,
1

Is there any way to train mmj_genai model with our custom dataset in Nvidia metropolis services? I’m following this page

2

I will check and feedback later.

3

The model we use in the mmj_genai example is based on OWL-ViT from google. The OWL-ViT github repository has some resources on how to train it. scenic/scenic/projects/owl_vit at main · google-research/scenic · GitHub

4

I had gone through the model, try to setup environment but getting to much version compatibility error. I’ve tried with python 3.7, 3.8, 3.9, 3.10. But getting version compatibility issue. I’m using master branch of this git hub repository. Could you please suggest me which branch should I use with which python version to set my environment?

Thanks in advance!!!

6

Can you ask in the owner of OWL-ViT for model trainning issue? Thanks.

7

How can we convert this Owl-ViT model to .engine? As you are using an engine file in the GenAI application.

8

Can you submit new topic for new question for better tracking topics? Thanks.

1 Like
10

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.