Skip to content

Exception while auto-train with regression task #363

Closed

Description

@smigielns

Hello

I would like to report a below error. I have tried also to run it using Visual Studio and ML.NET Model Builder but the result was the same.

obraz

debug_log.txt

Activity

  1. LittleLittleCloud commented on Nov 22, 2019

    @LittleLittleCloud
    Member

    Can we have your datasets, it should be something wrong in automl

  2. smigielns commented on Nov 25, 2019

    @smigielns
    Author

    Yes, of course.

    "Price" is a column to predict.
    testujemy.zip

    I'm not familiar with ML. I have ran the test just to see how it works so data file may be also not correct.

    I hope that helps.

  3. LittleLittleCloud commented on Dec 2, 2019

    @LittleLittleCloud
    Member

    I've run the experiment for several time and can't reproduce the error... Could you also try re-run that experiment and see if that error occur?
    and BTW could I have the version of your mlnet cli, and the test file ".\test_walidaca..csv"?
    Thanks!

  4. smigielns commented on Dec 9, 2019

    @smigielns
    Author

    test_walidacja.zip

    My mlnet version is:
    0.15.28007.4 @BuiltBy: dlab14-DDVSOWINAGE054 @Branch: features/automl @srccode: https://github.com/dotnet/machinelearning/tree/dc9a9b7ffcaf636541fe997c59f3bfdda57501e5+dc9a9b7ffcaf636541fe997c59f3bfdda57501e5

    I have run it also using Visual Studio and the error occurs when I increase the 'Time to train' value. It stops after 8th iteration - the same as when I run it with CLI, the log file you can find in the first message in this conversation.

    In below example I used only 'testujemy.csv' file.

    ml_error

    Have a good day!

  5. JakeRadMSFT commented on Jan 3, 2020

    @JakeRadMSFT
    Member

    @smigielns thanks for the added information. We'll try to reproduce the issue.

  6. added this to the January 2020 milestone on Jan 4, 2020
  7. LittleLittleCloud commented on Jan 7, 2020

    @LittleLittleCloud
    Member

    I can reproduce the bug you encountered now

  8. LittleLittleCloud commented on Jan 7, 2020

    @LittleLittleCloud
    Member

    It's a bug in AutoML Sweepable parameters, in SweepableParamAttributes, if valueText is "0,034456" (represented in Russian format) or something, it will be parsed into 34456 instead of 0.034456, If that parameter is significant ( like learning_rate), such error will cause unexpected training result ( r-square be -inf). But it won't break pipeline because it's not an exception. But in the next handling training result step, it will throw an exception because there's no legitimate training result and the bestFoldIndex will be -1.

  9. LittleLittleCloud commented on Jan 7, 2020

    @LittleLittleCloud
    Member

    This will be fixed in the next release, regards

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions