pp.mergecand( candata )
timer.send('Reduce LSH')
#candstats(dataset , bic.getinfo(agdataO))
# generate biclusters
link( candata, bic.genbic, bicdata, (bic.getinfo(agdataO), probabilityO, thr, (min_rows,min_cols), sparse, True), append=True )
timer.send('Gen. Bicluster')
'''
timer.close()
#pp.filterbics(bicdata)
pp.merge(dataset)
#pp.hierclust(dataset,7)
pp.uncovered(dataset, bic.getinfo(agdataO))
ev.stats(dataset) # print some results statistics
ev.microprecision(dataset) # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset)
ev.PMI(dataset)
#net.genetwork(dataset) # generate a network of objects and features
#ev.stats(dataset,'LSH') # print some results statistics
#ev.microprecision(dataset,'LSH') # calculate the microprecision if the objects have class embedded on their names
#ev.NMI(dataset,'LSH')
#ev.PMI(dataset,'LSH')
if __name__ == "__main__":
# HEY LISTEN!
# FILTER BICLUSTERS BY SORTING THEM AND SELECTING
# THOSE THAT COVER EVERYTHING FOUND
# generate biclusters
link(candata,
bic.genbic,
bicdata, (bic.getinfo(agdataO), probabilityO, thr,
(min_rows, min_cols), sparse, True),
append=True)
timer.send('Gen. Bicluster')
timer.close()
#pp.filterbics(bicdata)
pp.merge(dataset)
#pp.hierclust(dataset,7)
#pp.uncovered(dataset, bic.getinfo(agdataO))
ev.stats(dataset) # print some results statistics
ev.microprecision(
dataset
) # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset)
ev.PMI(dataset)
ev.stats(dataset, 'InClose') # print some results statistics
ev.microprecision(
dataset, 'InClose'
) # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset, 'InClose')
ev.PMI(dataset, 'InClose')
if __name__ == "__main__":
# generate biclusters
link(candata,
bic.genbic,
bicdata, (bic.getinfo(agdataO), probabilityO, thr,
(min_rows, min_cols), sparse, True),
append=True)
timer.send('Gen. Bicluster')
timer.close()
#pp.filterbics(bicdata)
pp.merge(dataset)
#pp.hierclust(dataset,7)
pp.uncovered(dataset, bic.getinfo(agdataO))
ev.stats(dataset) # print some results statistics
ev.microprecision(
dataset
) # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset)
ev.PMI(dataset)
ev.stats(dataset, 'LSH') # print some results statistics
ev.microprecision(
dataset, 'LSH'
) # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset, 'LSH')
ev.PMI(dataset, 'LSH')
if __name__ == "__main__":
pp.mergecand( candata )
timer.send('Reduce LSH')
#candstats(dataset , bic.getinfo(agdataO))
# generate biclusters
link( candata, bic.genbic, bicdata, (bic.getinfo(agdataO), probabilityO, thr, (min_rows,min_cols), sparse, True), append=True )
timer.send('Gen. Bicluster')
timer.close()
#pp.filterbics(bicdata)
pp.merge(dataset)
#pp.hierclust(dataset,7)
#pp.uncovered(dataset, bic.getinfo(agdataO))
ev.stats(dataset) # print some results statistics
ev.microprecision(dataset) # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset)
ev.PMI(dataset)
ev.stats(dataset, 'InClose') # print some results statistics
ev.microprecision(dataset, 'InClose') # calculate the microprecision if the objects have class embedded on their names
ev.NMI(dataset, 'InClose')
ev.PMI(dataset, 'InClose')
if __name__ == "__main__":
sparse = 1.0 # sparseness rate
print 'House Votes 84 '
dataset = 'house-votes-84' # dataset name
