Getting error while creating Responsible Ai dashboard about mlops-v2

[setuc]

Can you share which environment are you using? It's possible that the environment doesn't contain xgbosst. Consider adding that package as part of conda.yaml
@yogidosalwar

[yogidosalwar]

@setuc Here is my conda.yaml .
Thank you for quick response.

Here, i am stuck

[yogidosalwar]

@setuc here is screen shot of DevOps

Thank you.

[setuc]

Ahh. Got it. The responsbileAi modules user the environment that's different than the base container. It refers to the container from the Rai-vnext-Preview code. You can check the Rai component yaml file to see how the environment file is being built.

[yogidosalwar]

Ok. I will check.
Thank you.

[setuc]

Do you have your pipeline? Or did you just replace the model with XGBoost?

[yogidosalwar]

@setuc just i replace the model with XGBoost and added model dependencies in requirment.txt , train-conda.yml and train-monitor-conda.yml.

[yogidosalwar]

@setuc I went through the Rai-component.yml file and I made changes in requirment.txt which is part of Rai-vnext-Preview.
But still, getting error in the Run Azure ML pipeline. If you have any input, please let me know.

Thank you.

[cindyweng]

Can you see what's going on in the AML pipeline? Via the studio

[yogidosalwar]

@cindyweng @setuc I have RAI-vNext-Preview repo and MLOps-Test folder on my github. I made some changes as per my requirement in deploy-model-training-pipeline. yml, prep. py etc. and as well as I did changes in create_rai_insights.py. While running deploy-model-training-pipeline getting error while creating RAI Insights Dashboard Constructor. I went through the azure machine learning studio jobs and observe the code, but my changes are not showing there.

i did changes here
parser.add_argument("--maximum_rows_for_test_dataset", type=int, default=5000)
to
parser.add_argument("--maximum_rows_for_test_dataset", type=int, default=10000)

below changes i did in deploy-model-training-pipeline.yml

    - repository: RAI-vNext-Preview  # Template Repo  #rai-vnext-preview
      name: yogidosalwar/RAI-vNext-Preview # need to change org name from "Azure" to your own org #rai-vnext-preview
      endpoint: github-connection # need to set up and hardcode
      type: github
      ref: main  

Thank you

[yogidosalwar]

@setuc @cindyweng Hi any update on above issue or suggestion???
Thank you