pyphi.campaign.runner.partition_stride_entries#
- pyphi.campaign.runner.partition_stride_entries(system, direction, mechanism, purview, parts, indices, scheme)[source]#
Build one partition-stride cell’s payloads, one entry per pin.
φ per specified-state pin is a minimum over partitions; pin selection is a maximum over pins. The stride must therefore report every pin’s local minimum — not only the pins that win locally — so the merge can take the cross-stride minimum per pin before selecting. The pin enumeration is partition-independent, so every stride reports the same pin set. Pin-less formalisms (e.g. IIT 3.0) report the single plain minimum over the stride’s partitions.