My model worked in notebooks, where I just pip installed the required packages. But I am trying to run on a pipeline, and when I submit the pipeline as a job I get stuck in the Preparing phase. I believe this is due to the environment not resolving correctly. I've attached my .yml files, and I'm wondering where the issue might be. conda_depencies.yml name: pytorch-env-with-optuna channels: - conda-forge dependencies: - python=3.8 - numpy - pandas - scikit-learn - pip - pip: - torch - optuna - mltable - azure-ai-ml - azure-identity component.yml $schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json name: display_name: version: 6.0.0 type: command description: > inputs: outputs: code: . environment: image: mcr.microsoft.com/azureml/curated/acpt-pytorch-1.11-py38-cuda11.3-gpu:9 conda_file: conda_dependencies.yml command: > I've also tried removing optuna from the dependencies list just so I can get past the preparing phase, and that didn't work either. I'm not sure what I'm doing wrong. Continue reading...
