Description
Return clipped sample weight and its lost share of original total.
Weights, hash collisions and regression diagnostics have estimator-specific meanings. A local count, signed contribution or residual scaling can be calculated from supplied inputs without establishing sampling validity or model quality.
Return clipped sample weight and its lost share of original total.
When to use this tool
Use Sample Weight Clipping Mass Loss when checking the stated weight, hash or diagnostic contribution.
How to use the tool and how it works
- Enter weight, cap, totalWeight. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read clipped, lostMass, lostTotalFraction. Use the output convention below.
Return clipped sample weight and its lost share of original total. Require nonnegative weight/cap and positive totalWeight >= weight; clipped = min(weight,cap), lostMass = max(0,weight-cap), lostTotalFraction = lostMass/total.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| weight | Weight | Required. Finite number; minimum 0. |
| cap | Cap | Required. Finite number; minimum 0. |
| totalWeight | Total Weight | Required. Finite number; strictly above 0. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| clipped | Clipped | Required. Finite number. |
| lostMass | Lost Mass | Required. Finite number. |
| lostTotalFraction | Lost Total Fraction | Required. Finite number. |
{
"input": {
"weight": 12,
"cap": 5,
"totalWeight": 100
},
"output": {
"clipped": 5,
"lostMass": 7,
"lostTotalFraction": 0.07
}
}Limitations and assumptions
- The denominator is the original total; no reweighting, normalization or estimator impact is computed.
- 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 a complete estimator or feature pipeline and justified assumptions to assess statistical impact, uncertainty and model behavior.
References
Similar or alternative tools
- Sample Weight Unknown Support Flag
Separate known and unknown sample-weight totals.
- Bootstrap One Draw Multiplicity Term
Return one observation's multiplicity and share in supplied bootstrap draws.