## BUILDING UP DENSITY
print "computing density"
maxi = 0
for key in brands.keys():
if brands[key] > maxi:
maxi = brands[key]
n_key = key
print "key : %s, number : %s" % (n_key,brands[n_key])
density = compute_density(brands)
##################################################################################
## PRINTING STUFF
for i in xrange(1,15):
try:
print "number of items : %s, number of brands : %s" % (i, density[i])
except:
pass
print "number of brands : %s" % (len(brands.keys()),)
print "number of items : %s" % (count)
print "ratio items/brands : %s" % ((count-none_number)/float(len(brands.keys())))
print "number of un_brand items : %s" % (none_number,)
smart_plot(map(lambda x : density[x],density.keys()),x_list=sorted(density.keys()))
break
spam_reader = parser(file_name)
leng = int(prix_max / pas_interval) + int(
bool(prix_max / pas_interval - int(prix_max / pas_interval))) + 1
l = [0] * leng
#print prix_max
##remplir la liste l avec le nombre d'objet de categorie cat pour chaque intervalle
count = 0
for row in spam_reader:
if row[3] == cat:
#print int(float(row[8])/pas_interval)
l[int(float(row[8]) / pas_interval)] += 1
count += 1
if count == limit:
break
##tracer la densité empirique pour la categorie cat
g = range(leng)
def foo(x):
return x * pas_interval
g = map(foo, g)
smart_plot(l, g)
for row in spam_reader:
if row[3] == cat:
#print int(float(row[8])/pas_interval)
l[int(float(row[8])/pas_interval)]+=1
count += 1
if count == limit:
break
##tracer la densité empirique pour la categorie cat
g=range(leng)
def foo(x):
return x*pas_interval
g=map(foo,g)
smart_plot(l,g)
##################################################################################
## BUILDING UP DENSITY
print "computing density"
maxi = 0
for key in brands.keys():
if brands[key] > maxi:
maxi = brands[key]
n_key = key
print "key : %s, number : %s" % (n_key, brands[n_key])
density = compute_density(brands)
##################################################################################
## PRINTING STUFF
for i in xrange(1, 15):
try:
print "number of items : %s, number of brands : %s" % (i, density[i])
except:
pass
print "number of brands : %s" % (len(brands.keys()), )
print "number of items : %s" % (count)
print "ratio items/brands : %s" % (
(count - none_number) / float(len(brands.keys())))
print "number of un_brand items : %s" % (none_number, )
smart_plot(map(lambda x: density[x], density.keys()),
x_list=sorted(density.keys()))
