Open TillF opened 7 years ago
Significantly improved with branch large_prof_class (#30 ), but there are still some redundancies that could be eliminated (marked in the code with ii). Added some suggestions to the user in the FAQ (wiki).
We could also exploit the rstats_head.txt better: currently, setting 0 or 1 for an attribute means the same: Do not make any clustering on it. It would be nice to have 1: do not make any clustering on this attribute, but keep it in the output. 0: completely ignore this attribute. This would allow removing attributes even after area2catena() if it runs out that they consume to much memory.
The modified version of prof_class of branch large_prof_class does not yet correctly treat the argument EHA_SUBSET.
prof_class() runs out of memory with large files of rstats.txt. This has been partially reported in #16