Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 19: Interpret Model (Feature Impacts on Target (Overall Impact: a…
Chapter 19: Interpret Model
Feature Impacts on Target
Overall Impact: a feature without consideration of the impact of other features
The overall impact on a feature adjusted for the impact of other feature
The directional impact of a feature
The partial impact of a feature
The overall impact of a feature Adjusted for the impact of other features
Click on the top model as sorted by cross validation, and the select Feature Impact, followed by Compute Feature Impact
Doing this will initiate a calculation of the value of each feature in the context of this model
DataRobot does this by randomly shuffling the values of one feature in the context of this model
after this process is carried out, datarobot produces the view
The third type of relationship is what has been termed in this book the directional impact of the feature, or whether the presence of a value helps the model by assisting it in predicting readmissions or non-readmissions
While the datarobot Variable Effect screen does no provide all the information to recreate a model, it does provide what are commonly known as coefficients
As you get more comfortable, consider looking at the leaderboard for any directional impact information
Model X Ray: constructs a list of features ranked by their influence on the target as denoted by the size of the green line under each feature as it is processing the rest
Filter Stop Words: A stop word ic a common word generally assumed to have little value
Hot spots: In Insights. Shows the most relevant (up to four) combination of features and their effect on the target
Hot/Cold column header: to select only hot (red-readmissions) or cold (blue-non-readmissions) sub-groups
The reason codes are a powerful feature that can be supplement business decisons
computing reason codes is slower than computing predictions