Using Azure Automated ML

In the Azure ML studio, create a new dataset using the taxi fare data file from Microsoft.

Change the timeout of the job to 15 minutes by expanding limits on the task settings page, and set the Experiment timeout (minutes) to 15, so it will stop after approximately 15 minutes.

Take a look at some of the metrics for the run, like the R2 score, or the root mean squared error.

Make sure you change the virtual machine to a DS2_v2 size and change the instance count to 1, or your resources will be deleted.

• Use the test tab and input the JSON data below to test your model and see how accurate the predicted value is compared to the actual value of 17.

  {
     "input_data":{
        "data":[
           {
              "vendor_id":"CMT",
              "rate_code":1,
              "passenger_count":1,
              "trip_time_in_secs":1271,
              "trip_distance":3.8,
              "payment_type":"CRD"
           }
        ]
     }
  }