Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapters 21-23 (Chapter 23: Create Model Monitoring and Maintenance Plan…
Chapters 21-23
Chapter 23: Create Model Monitoring and Maintenance Plan
23.1 Potential Problems
need to create a monitoring and maintenance plan
Perhaps more difficult to detect is the fact that the ailments of diabetes patients may change in a positive direction
can result from new medications or detection of a deleterious effects from the interaction effects a new medication with other medications
23.2 Strategies
DataRobot will fail rather than attempt to make the best of the available data
if a column is removed from the prediction data, the system fails
Regardless of the reason for deteriorating model performance, early detection of is paramount
Chapter 21: Choose Deployment Strategy
DataRobot provides a number of different deployment strategies
DataRobot Prime
creates an approximation of your selected model available as code in the Python and Java programming languages, and its availability to you is based on your DataRobot account type
Batch
uses the DataRobot API to upload and score multiple large files in parallel
Application Programming Interface (API)
An Application Programming Interface (API) is created on the DataRobot server allowing a programmer to write a program that uploads data for a new patient to this API and receive back a probability
In-place with Spark
allows exporting the selected model as an executable file to be used in an Apache Spark environment
Spark is a fast and massively distributed data processing environment
Drag-and-drop
doesn't need to include a target since this is what we're trying to predict
include at least the reference id column to avoid having to match the results based on the row id provided by DataRobot
upload data in files containing all the features used to create the model
Chapter 22: Document Modeling Process for Reproducibility
22.1 Model Documentation
Work under the assumption that you will be back to this project within a year
what info would be helpful for you to have immediate access to
Assuming such a system is in place, our next step is to specify the business problem
Proper documentation is important for others to understand what was done to accomplish the results and the justification for the project to exist
Next, we document where the data came from
Documenting the modeling process is where projects most often fail
The next step is to specify all the steps taken in DataRobot
Finally, we record the business rules for use of the model and the probability thresholds along with information from the IT group about how the model was set up