A vectorized bootstrap runs out of memory. Can you reduce batch size without reducing its total resample count?
StatisticsLevel: IntermediateBootstrap uncertainty
A vectorized bootstrap runs out of memory. Can you reduce batch size without reducing its total resample count?
Hints
- Separate total work from peak working set.
- The batch parameter controls concurrent resamples.
Answer
Yes. Process fewer resamples per call while preserving the total count. Working memory scales with batch×sample size, although stored results still consume memory.
Source and licence: SciPy — resampling and Monte Carlo inference · BSD-3-Clause · answer: Quantoros editorial