topic What are the parameters For MLflow Project file in Machine Learning

What are the parameters For MLflow Project file

confusedIntern — Tue, 12 Jul 2022 18:08:47 GMT

Hi! I was just wondering what are the parameters For MLflow Project file?

I'm following this tutorial to create my own MLflow Project: https://docs.databricks.com/applications/mlflow/projects.html

and within this tutorial, the MLproject file looks like this:

name: My Project
 
conda_env: my_env.yaml
 
entry_points:
  main:
    parameters:
      data_file: path
      regularization: {type: float, default: 0.1}
    command: "python train.py -r {regularization} {data_file}"
  validate:
    parameters:
      data_file: path
    command: "python validate.py {data_file}"

How did they determine what the parameters are? What exactly are parameters? Are they the parameters from when you do log_param in Mlflow in the notebook?

Please help! Many thanks!

Re: What are the parameters For MLflow Project file

Prabakar — Sun, 24 Jul 2022 21:04:47 GMT

You can refer to MLflow doc for more details.

Re: What are the parameters For MLflow Project file

Vidula — Sun, 04 Sep 2022 05:28:18 GMT

Hi there @Margie Kale

Does @Prabakar Ammeappin response answer your question? If yes, would you be happy to mark it as best so that other members can find the solution more quickly?

We'd love to hear from you.

Thanks!

Re: What are the parameters For MLflow Project file

Anonymous — Mon, 05 Sep 2022 05:13:50 GMT

I am also encountering this issue. Don't know why its happening.

Re: What are the parameters For MLflow Project file

sean_owen — Mon, 19 Sep 2022 22:32:40 GMT

These parameters are parameters that you will specify when you run the MLflow Project with the mlflow CLI. It lets you parameterize your code, and then pass different parameters to it. How you use them is up to your code. These are not model hyperparameters, of the type you might log with a model.