Description
Flag a category absent from a supplied training-category list.
Data workflows require explicit training scope, feature/label availability, missingness codes, group assignments and evaluation direction. Supplied training statistics are not proof that preprocessing avoided leakage. A local target-encoding term or retransformed prediction depends on assumptions that must be checked in the complete model and validation design.
Flag a category absent from a supplied training-category list.
When to use this tool
Use Feature Value Unseen Category Flag when you already have category, trainingCategories and need to flag a category absent from a supplied training-category list.
How to use the tool and how it works
- Enter category, trainingCategories. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read unseen. Inspect unseen before interpreting the associated values.
Return !trainingCategories.includes(category) using literal equality.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| category | Category | Required. Text; maximum UTF-16 length 8192. |
| trainingCategories | Training Categories | Required. Array of text; maximum UTF-16 length 8192; maximum items 256. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| unseen | Literal category absence; an empty training list makes this true for any category. | Required. Boolean. |
{
"input": {
"category": "new",
"trainingCategories": [
"a",
"b"
]
},
"output": {
"unseen": true
}
}Limitations and assumptions
- Empty strings and duplicates are accepted; an empty list marks every category unseen. The list is not fitted or verified as training-only, and no fallback encoding is selected.
- Inputs must satisfy both the field types and the cross-field conditions above. Invalid input causes an error; a successful false flag or documented placeholder is a different result.
Alternative or Complementary methods
Review the full data lineage, preprocessing fit, split design and evaluation procedure for model conclusions. Estimate required statistics only within the intended training folds and verify actual availability; local supplied-value transformations cannot establish those properties by themselves.
References
Similar or alternative tools
- One Hot Reference Level Drop Flag
Flag literal equality with a supplied one-hot reference category.
- Train Test Single Identifier Collision
Check whether one supplied identifier appears in both training and test lists.