search

eXascaleInfolab / PyCABeM

Python Benchmarking Framework for the Clustering Algorithms Evaluation: networks generation and shuffling; failover execution and resource consumption tracing (peak RAM RSS, CPU, ...); evaluation of Modularity, conductance, NMI and F1 Score for overlapping communities
Other
19 stars 4 forks source link

NMI_s results for multiple sources are duplicated #12

Open luav opened 8 years ago

luav commented 8 years ago

This issue exists when the benchmark is executed on multiple (2 in the example below) directories and exists for NMI_s only (not for NMI or mod). Example of oslom2 50K10.nmi_s, but it is the same for all algorithms:

# NMI_s level[/shuffle]
0.0632715   tp2
0.158837    tp1
0.549299    tp
0.0632715   tp2
0.158837    tp1
0.549299    tp

This issue could be caused by the benchmark interruption and reexecution, which should be checked and clarified.

Either backup on reexecution should be fixed, or paths traversal for multiple source dirs.