Description
Validate both probabilities in [0, 1].
Negative log probability scores an observed event. Flooring probabilities changes that score and support rules distinguish zero contribution from an infinite mismatch.
Validate both probabilities in [0, 1].
When to use this tool
Use Cross Entropy Reference Zero Support Flag when you need one explicit log-probability policy or support diagnostic.
How to use the tool and how it works
- Enter sourceProbability, referenceProbability. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read infinite, zeroContribution. Inspect infinite, zeroContribution before interpreting the associated values.
Validate both probabilities in [0, 1]. Mark infinite for positive sourceProbability with zero referenceProbability and zeroContribution whenever sourceProbability is zero.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| sourceProbability | Source Probability | Required. Finite number; minimum 0; maximum 1. |
| referenceProbability | Reference Probability | Required. Finite number; minimum 0; maximum 1. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| infinite | Infinite | Required. Boolean. |
| zeroContribution | Zero Contribution | Required. Boolean. |
{
"input": {
"sourceProbability": 0.25,
"referenceProbability": 0
},
"output": {
"infinite": true,
"zeroContribution": false
}
}Limitations and assumptions
- This flags support cases without computing a logarithmic contribution; full distributions and normalization are not checked.
- 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
Keep clipping and normalization policies consistent across the complete probability model and evaluation set.
References
Similar or alternative tools
- Information Gain Zero Cell Limit
Return cell × (ln(cell) - ln(reference)), with zero at a zero cell and zeroCellLimit flagged.
- Information: Relative Entropy Single Cell Term
Return p × (ln(p) - ln(q))/ln(base), or zero for p zero.