ENDPOINTS.md

Python functions

The modeling project must provide a Python function for each of the following tasks:

• Package Defaults: Get the default Model Parameters and their meta data.
• Parse user adjustments: Do model-specific formatting and validation on user adjustments.
• Run simulation: Submit the user adjustments (or none) to the model to run the simulation.

Package defaults

Accepts Meta Parameters, if they are being utilized. Returns data in the form specified in the "Model Parameters" section of IOSCHEMA.md.

• Python:

  import matchups
  
  def package_defaults(**meta_parameters):
      return matchups.get_inputs(use_full_data=meta_parameters["use_full_data"])

  Here's what you get after filling in this function:

Parse user adjustemnts

Accepts parsed user adjustments, separated by each major section. Returns parsed user adjustments, JSON representation of the user adjustments, and warnings/errors (if any).

Compute Studio will provide parsed user adjustments of the form:

{
    "matchup": {
        "start_date": [{"value": "2012-08-01"}],
        "end_date": [{"value": "2012-09-01"}],
        "pitcher": [{"value": "Not a Real Pitcher"}],
    }
}

The function should return:

Potentially, reformatted data (no reformatting necessary for this project):

{
    "matchup": {
        "start_date": [{"value": "2012-08-01"}],
        "end_date": [{"value": "2012-09-01"}],
        "pitcher": [{"value": "Not a Real Pitcher"}],
    }
}

JSON representation of each major section of parameters:

    {
        "matchup": '{"start_date": [{"value": "2012-08-01"}], "end_date": [{"value": "2012-09-01"}], "pitcher": [{"value": "Not a Real Pitcher"}]}'
    }

Warnings/Errors:

    {
        "matchup": {
            "errors": {
                "pitcher": ["Pitcher \"Not a Real Pitcher\" not allowed"]
            },
            "warnings": {}
        }
    }

• Python:

  import matchups
  
  def parse_user_inputs(params, jsonparams, errors_warnings,
                          **meta_parameters):
      # parse the params, jsonparams, and errors_warnings further
      use_full_data = meta_parameters["use_full_data"]
      params, jsonparams, errors_warnings = matchups.parse_inputs(
          params, jsonparams, errors_warnings, use_full_data==use_full_data)
      return params, jsonparams, errors_warnings

  Here's what you get after filling in this function:

  The jsonparams data is displayed on the outputs page:

  

Run simulation

Accepts Meta Parameters values and parsed and formatted user adjustments. Returns outputs as specified by the Outputs schema

Compute Studio submits the model's meta parameters and the parsed and formatted user adjustments:

    {
        "meta_parameter1": value,
        "meta_parameter2": value,
        ...
        "user_mods": {
            "major_section1": {
            ...
            },
            "major_section2": {
            ...
            },
            ...
        }
    }

The function returns the results of the simulation:

Outputs schema

• Python:

  import matchups
  
  def run(**kwargs):
      result = matchups.get_matchup(kwargs["use_full_data"], kwargs["user_mods"])
      return result

  Here's what you get after filling in this function:

  Aggregate outputs:

  Outputs:

  Toggle along dimension: