az feedback auto-generates most of the information requested below, as of CLI version 2.0.62
Describe the bug
Details are in microsoft/MLOpsPython#136
To Reproduce
NOTE: to repro the error reported in microsoft/MLOpsPython#136 you will need more setup, such as creating ML workspace, register a model etc. The following repro step is simplified. You won't see an error but you will see the symptom of different version SDK is loaded.
step 1: install latest azure-cli-ml: az extension add -n azure-cli-ml (or az extension update -n azure-cli-ml).
You should get version 1.0.85. SDK version is correct as you can see from
$HOME/.azure/cliextensions/azure-cli-ml/azureml_core*
step 2: create a pyenv, can be conda, venv, doesn't matter. I use venv
python -m venv test-venv
source test-venv/bin/activate
step 3: install a different version of azureml-core
pip install azureml-core==1.0.79
step 4: run an ml CLI command: az ml workspace list -g abc --debug (doesn't matter whether resource group abc exists or not)
In the debug level log, you will see SDK 1.0.79 is loaded (you will see 'azureml-core 1.0.79' in debug output)
In the real error case, there is a change in azureml-core 1.0.81 which introduces a new required parameter.
Expected behavior
A clear and concise description of what you expected to happen.
Environment summary
Install Method (e.g. pip, interactive script, apt-get, Docker, MSI, edge build) / CLI version (az --version) / OS version / Shell Type (e.g. bash, cmd.exe, Bash on Windows)
Additional context
Add any other context about the problem here.
Describe the bug
Details are in microsoft/MLOpsPython#136
To Reproduce
NOTE: to repro the error reported in microsoft/MLOpsPython#136 you will need more setup, such as creating ML workspace, register a model etc. The following repro step is simplified. You won't see an error but you will see the symptom of different version SDK is loaded.
step 1: install latest azure-cli-ml: az extension add -n azure-cli-ml (or az extension update -n azure-cli-ml).
You should get version 1.0.85. SDK version is correct as you can see from
$HOME/.azure/cliextensions/azure-cli-ml/azureml_core*
step 2: create a pyenv, can be conda, venv, doesn't matter. I use venv
python -m venv test-venv
source test-venv/bin/activate
step 3: install a different version of azureml-core
pip install azureml-core==1.0.79
step 4: run an ml CLI command: az ml workspace list -g abc --debug (doesn't matter whether resource group abc exists or not)
In the debug level log, you will see SDK 1.0.79 is loaded (you will see 'azureml-core 1.0.79' in debug output)
In the real error case, there is a change in azureml-core 1.0.81 which introduces a new required parameter.
Expected behavior
A clear and concise description of what you expected to happen.
Environment summary
Install Method (e.g. pip, interactive script, apt-get, Docker, MSI, edge build) / CLI version (
az --version) / OS version / Shell Type (e.g. bash, cmd.exe, Bash on Windows)Additional context
Add any other context about the problem here.