Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 21,22,23 (Document Modeling Process For Reproducibility (Model…
Chapter 21,22,23
Document Modeling Process For Reproducibility
Throughout documentation allows you to focus on what you care about and learn more details about data used
Good notes save time in the future and allow you to prepare for a presentation easily
Model Documentation
Proper documentation is necessary to keep track of data outputs
Set up Prediction System
Choose deployment strategy
Strategies based on skill
DataRobot Prime
Batch
Application Programming inteface (API)
In-place with Spark
Drag-and-drop
Retraining Model
improve the predicitve ability
Ulock the holdout and put training data to 100%
Create Model Monitoring and Maintenance Plan
Potential Problems
Monitoring documentation is needed to be able to make changes and let other people understand what data means
It will benefit the model effectiveness
Strategies
Always run model when new data becomes available DataRobot helps us with this
You can analyze model deterieorizing by evaluating model performance