Description
Look up a categorical probability while distinguishing a missing label.
Probability calculations distinguish event mass, density likelihood, support membership and interval inclusion. Zero likelihood is not always proof of impossibility, and a known zero-probability category differs from an absent label. Independence, disjointness and supplied CDF values are assumptions or records, rather than properties established by local arithmetic.
Look up a categorical probability while distinguishing a missing label.
When to use this tool
Use Categorical Label Probability Lookup when you already have label, labels, probabilities and need to look up a categorical probability while distinguishing a missing label.
How to use the tool and how it works
- Enter label, labels, probabilities. Follow any array-alignment requirements below. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read found, probability. Inspect found before interpreting the associated values.
Require aligned arrays, unique labels and probabilities in [0,1] summing to one within EPSILON*length. Return found and the matching probability, or zero when missing.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| label | Label | Required. Text; maximum UTF-16 length 8192. |
| labels | Labels | Required. Array of text; maximum UTF-16 length 8192; maximum items 256. |
| probabilities | Probabilities | Required. Array of finite number; maximum items 256. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| found | Distinguishes a present zero-probability label from an absent label with probability=0 placeholder. | Required. Boolean. |
| probability | Probability | Required. Finite number. |
{
"input": {
"label": "B",
"labels": [
"A",
"B"
],
"probabilities": [
0.25,
0.75
]
},
"output": {
"found": true,
"probability": 0.75
}
}Limitations and assumptions
- A found zero-probability label differs from found=false; labels are literal and the list is not fitted to observations or renormalized.
- 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.
- Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Numeric signed-zero results are exported as zero.
Alternative or Complementary methods
Use the complete joint model when independence or disjointness cannot be assumed. Evaluate a distribution on its actual support for CDF or likelihood questions, and keep absent categories distinct from explicitly listed zero-probability outcomes.
References
Similar or alternative tools
- Likelihood Impossible Support Flag
Expose zero likelihood and a contradiction with supplied support membership.
- Finite Event Mass Remaining Budget
Calculate unassigned probability mass after supplied disjoint event masses.