def _get_id_stats(glyphs, k=None):
import stats
if len(glyphs) < 3:
return (len(glyphs),1.0, 1.0, 1.0)
if k is None:
k = kNN()
distances = k.unique_distances(glyphs)
return (len(glyphs),stats.lmean(distances), stats.lstdev(distances), stats.lmedian(distances))
def _get_id_stats(glyphs, k=None):
import stats
if len(glyphs) < 3:
return (len(glyphs),1.0, 1.0, 1.0)
if k is None:
k = kNN()
distances = k.unique_distances(glyphs)
return (len(glyphs),stats.lmean(distances), stats.lstdev(distances), stats.lmedian(distances))
def characteristic_value(OD_list):
if characteristic_value_type == 'median':
return float('%1.4f'%(stats.lmedian(OD_list)))
elif characteristic_value_type == 'max':
return float('%1.4f'%(max(OD_list)))
else:
print 'characteristic value type not defined, enter median or max'
exit()
def characteristic_value(plate_data):
for keys in plate_data:
if not characteristic_value_of_time_course.has_key(keys):
if characteristic_value_type == 'median':
characteristic_value_of_time_course[keys] = stats.lmedian(plate_data[keys])
else:
characteristic_value_of_time_course[keys] = max(plate_data[keys])
for keys in characteristic_value_of_time_course:
print keys, characteristic_value_of_time_course[keys]
rdp_list = []
rdp_tmp = 0.0
if icmp_len < bn_len :
loop_count = icmp_len
for i in range(loop_count) :
if brunet_time_list[i] != -1.0 and icmp_time_list[i] != -1.0 :
if brunet_time_list[i] != 0.0 and icmp_time_list[i] != 0.0 :
bin_index = math.floor(icmp_time_list[i]/binsize)*binsize
rdp_tmp = brunet_time_list[i]/icmp_time_list[i]
tmp1_list = []
if bin_index in timebin_to_time_list :
tmp1_list = timebin_to_time_list[bin_index]
tmp1_list.append(rdp_tmp)
else :
timebin_to_time_list[bin_index] = [rdp_tmp]
timebin_to_time_list[bin_index] = tmp1_list
#print timebin_to_time_list
bin_final_list = timebin_to_time_list.keys()
bin_final_list.sort()
for bin in bin_final_list :
if len(timebin_to_time_list[bin]) > 0:
tmed = stats.lmedian(timebin_to_time_list[bin],100000 )
tmp_scor_list = timebin_to_time_list[bin]
tmp_scor_list.sort()
nin = int( math.floor( 0.9*len(tmp_scor_list) ) )
tstdband = tmp_scor_list[nin]
print bin, min(tmp_scor_list),tmed ,tstdband
def median(self):
return float('%1.4f'%stats.lmedian(self.datapoints))