search

microsoft / vscode-tools-for-ai

Azure Machine Learning for Visual Studio Code, previously called Visual Studio Code Tools for AI, is an extension to easily build, train, and deploy machine learning models to the cloud or the edge with Azure Machine Learning service.
Other
325 stars 95 forks source link

No way of seeing what is wrong with a failed job #83

Closed syagev closed 5 years ago

syagev commented 5 years ago

VSCode 1.28.2 VSCode-tools-for-AI 0.3.1 Windows 10/1809

When submitting a job using docker:

{
    "version": "0.0.1",
    "job": {
        "type": "General",
        "files": {
            "includes": [
                "**/*.py"
            ]
        },
        "name": "Config",
        "startupCommand": "python",
        "startupScript": "simu_mcmk_get_config_v2.py",
        "arguments": "-opdir /workspace",
        "workingDirectory": "/code",
        "batchArguments": [],
        "docker": {
            "useRoot": false,
            "type": "Docker",
            "image": "gcr-repos.redmond.corp.microsoft.com:5000/styagev/simulation_multisource_v2"
        },
        "author": "MIDDLEEAST.styagev"
    },
    "platform": {}
}

my job fails (get's "interrupted"). There is no information in the job detail for me to see what went wrong.

image

Also - I couldn't find in the docs a simple explanation of what is happening when I "submit a job" to a linux remote machine. Is the code copied and mounted on the container? where? Can I mount additional data directories on the container?

Thanks

greazer commented 5 years ago

This extension has been redesigned to target scenarios only dealing with Azure Machine Learning. In the archive folder, you can find the 0.3.1 vsix installer that includes previous features before this redesign, along with corresponding docs.