Forum Discussion

jing_you's avatar
jing_you
Regular Visitor
2 years ago
Solved

AbortErrror for Experiment in Fabric

Hi,

I have cretaed a Experiment in Fabric but when i attempt to open it, getting error as followed. What is the cause and how to solve it?

 

data science
experiment
  • jing_you's avatar
    jing_you
    2 years ago

    I create a new experiment in group workspabce by code as follow on 24/7:

    EXPERIMENT_NAME = "experiment1"
    mlflow.set_experiment(EXPERIMENT_NAME)

    with mlflow.start_run(run_name = f'{model_select}_{i}') as run:
        model1.fit(X, y)
        y_pred1 = model1.predict(X)

        mse1 = mean_squared_error(y, y_pred1)
        r21 = r2_score(y, y_pred1)

        mlflow.log_metric('mse_linear', mse1)
        mlflow.log_metric('r2_linear', r21)

        signature = infer_signature(X, y)
        mlflow.sklearn.log_model(model, "model", signature = signature)

    But it cannot open and pop out the AbortError the next day. Then I try to create another new experiment, it also appear same issue. The workspace assign with activated 'fabric' capacity. 
    However, the experiment able to open today (29/7), is this just a bug or related to my operations or capacity?

3 Replies

  • jing_you's avatar
    jing_you
    Regular Visitor
    2 years ago

    I create a new experiment in group workspabce by code as follow on 24/7:

    EXPERIMENT_NAME = "experiment1"
    mlflow.set_experiment(EXPERIMENT_NAME)

    with mlflow.start_run(run_name = f'{model_select}_{i}') as run:
        model1.fit(X, y)
        y_pred1 = model1.predict(X)

        mse1 = mean_squared_error(y, y_pred1)
        r21 = r2_score(y, y_pred1)

        mlflow.log_metric('mse_linear', mse1)
        mlflow.log_metric('r2_linear', r21)

        signature = infer_signature(X, y)
        mlflow.sklearn.log_model(model, "model", signature = signature)

    But it cannot open and pop out the AbortError the next day. Then I try to create another new experiment, it also appear same issue. The workspace assign with activated 'fabric' capacity. 
    However, the experiment able to open today (29/7), is this just a bug or related to my operations or capacity?
    • Anonymous's avatar
      Anonymous
      Not applicable
      2 years ago

      Hi jing_you ,

      AFAIK, the fabric products required hosted on an active fabric capacity, if the capacity paused, the product also not able to be used.
      I'd like to suggest you check the fabric capacity status if it any change applied to it at the day the abort error occurred.

      In addition, which level of fabric capacity are you worked? I think you also need to confirm if it exists enough resource to execute these processing.

      Regards,

      Xiaoxin Sheng

  • Anonymous's avatar
    Anonymous
    Not applicable
    2 years ago

    HI jing_you,

    It seems like you only share less data for troubleshoot.
    Can you please share some more detail information about this issue? They should help us clarify your scenario and test to troubleshoot.
    For example:
    What experiment the issue appears? new created experiment or an old experiment that created in previous and works properly?
    Where the experiment host? group workspace or personal 'my workspace'.
    Did the workspace assign with activated 'fabric' capacity?
    What type of operations that you processed in notebook to create experiment?

    Regards,

    Xiaoxin Sheng