Description
Apply a supplied inclusive maximum-score threshold to explicit unit-sum probability rows, rejecting unresolved top ties rather than guessing a label; not a validated decision policy.
A confusion table uses actual labels as rows and predicted labels as columns. Cells can contain occurrence counts or supplied nonnegative weights; the axis label order must stay with the matrix. Row, column and joint normalization have different denominators, and a zero denominator needs an explicit undefined result.
Apply a supplied inclusive maximum-score threshold to explicit unit-sum probability rows, rejecting unresolved top ties rather than guessing a label; not a validated decision policy.
When to use this tool
Use this tool to apply a supplied inclusive maximum-score threshold to explicit unit-sum probability rows, rejecting unresolved top ties rather than guessing a label; not a validated decision policy.
How to use the tool and how it works
- Enter scores, labels, threshold using the input definitions below.
- Run the operation. Resolve malformed-input errors before interpreting the result.
- Read result.predictions. Interpret signs, units, ordering and null values using the rules below.
Validate score rows and a threshold in [0,1]. Accept a label only when the row maximum is at least threshold and exactly one label has that maximum; otherwise return label = null and rejected = true.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| scores | Score rows aligned with labels; every entry lies in [0,1] and each row sums to one within 10⁻⁸. | Dense list of at most 128 rows of at most 128 finite numbers; the operation’s shape constraints also apply. |
| labels | Distinct label strings defining actual row and predicted column order; score tools require at least one label. | Dense list of at most 128 strings, each at most 8,192 UTF-16 code units. |
| threshold | Inclusive maximum-score acceptance threshold from 0 through 1. | Finite number satisfying the stated comparisons. |
| Output field | Interpretation |
|---|---|
| result | Object containing the fields below. |
| result.predictions | Input-row prediction records retaining maximum score, ties or explicit rejection as stated above. |
{
"input": {
"scores": [
[
0.5,
0.5
],
[
0.1,
0.9
]
],
"labels": [
"a",
"b"
],
"threshold": 0.8
},
"output": {
"result": {
"predictions": [
{
"score": 0.5,
"label": null,
"rejected": true
},
{
"score": 0.9,
"label": "b",
"rejected": false
}
]
}
}
}Limitations and assumptions
- The supplied threshold is an explicit arithmetic policy, not a validated operating point for any real-world application.
- Label strings are compared exactly. Confusion values may be nonnegative fractional weights; a count interpretation requires count data. Probability-like score rows use [0,1] entries with sum tolerance 10⁻⁸, without establishing calibration.
- Every declared top-level input is required and unknown fields are rejected. Ordinary strings are limited to 8,192 UTF-16 code units and lists to 128 entries, with tighter operation-specific bounds where stated. JSON text has bounded nesting and node counts.
- Numeric outputs must be finite and within the absolute safe-integer magnitude; exact larger integers use decimal text. Source positions are zero-based UTF-16 offsets and span ends are exclusive unless a different convention is explicitly stated.
Alternative or Complementary methods
Check label coverage and axis order before comparing tables. Use a separate evaluation design for uncertainty, calibration, operating thresholds and the consequences of real-world decisions.
References
Similar or alternative tools
- Confusion Reject Column Partition
Partition one explicit predicted reject column without removing actual rows or relabeling rejected observations as errors.
- Prediction Top Label Tie Audit
Audit every maximum-probability label tie from explicit nonnegative unit-sum score rows; retain all ties and do not assume calibration or pick a hidden winner.