Re-using cached optimisation/integration results

Re-using cached optimisation/integration results#

For each process, Pepper stores optimisation/integration results under pepper_cache/{process}. This are usually written out the first time Pepper is run for that process. In subsequent runs, the results are read in, instead of calculating them again, which can take a significant amount of time.

Note

To disable writing cached results, you can use the setting write_cached_results = false. To ignore existing cached results, use use_cached_results = false. Both settings should be in the [main] section of your pepper.ini file.

When changing settings other than the actual process (e.g. the PDF being used), the stored results are still valid in the sense that the event generation gives correct results. However, the efficiency of the event generation might significantly deteriorate. This is in particular true for some more drastic changes, like switching from helicity sampling to helicity summing. In these cases, it might be preferable to remove the old results under pepper_cache/{process} and re-generate them in the next Pepper run.