def loadOptimized(self):
self.gc = GraphCollection()
self.qd1g_f = self.gc.load('resource/qd1g_f.pck')
self.qd2g_f = self.gc.load('resource/qd2g_f.pck')
self.qd3g_f = self.gc.load('resource/qd3g_f.pck')
self.length_f = self.gc.load('resource/lngt_f.pck')
self.qd1g_h = self.gc.load('resource/qd1g_h.pck')
self.qd2g_h = self.gc.load('resource/qd2g_h.pck')
self.qd3g_h = self.gc.load('resource/qd3g_h.pck')
self.length_h = self.gc.load('resource/lngt_h.pck')
def scanAndSave():
qd = GraphCollection()
fileName1 = 'resource/lemmad_utf.txt'
fileName2 = 'resource/lemmad_piir_utf.txt'
trie1 = qd.createTrie(fileName1, 3)
trie2 = qd.createTrie(fileName2, 3)
length1 = qd.createLengthQuickArr(fileName1)
length2 = qd.createLengthQuickArr(fileName2)
probability1 = qd.createQuickDict(trie1, 1)
probability2 = qd.createQuickDict(trie2, 1)
generatorD1 = qd.createQuickDict(trie1, 2)
generatorD2 = qd.createQuickDict(trie2, 2)
generatorT1 = qd.createQuickDict(trie1, 3)
generatorT2 = qd.createQuickDict(trie2, 3)
qd.save('resource/qd3g_f.pck', generatorT1)
qd.save('resource/qd2g_f.pck', generatorD1)
qd.save('resource/qd1g_f.pck', probability1)
qd.save('resource/lngt_f.pck', length1)
qd.save('resource/qd3g_h.pck', generatorT2)
qd.save('resource/qd2g_h.pck', generatorD2)
qd.save('resource/qd1g_h.pck', probability2)
qd.save('resource/lngt_h.pck', length2)
class Main:
gc = None
qd1g_f = None
qd2g_f = None
qd3g_f = None
length_f = None
qd1g_h = None
qd2g_h = None
qd3g_h = None
length_h = None
def run(self):
def avg(listObj):
return float(sum(listObj)) / len(listObj)
def median(listObj):
sorts = sorted(listObj)
length = len(sorts)
midIdx = int(length/2)
if not length % 2:
return (sorts[midIdx] + sorts[midIdx - 1]) / 2
return sorts[midIdx]
def mode(listObj):
from collections import Counter
data = Counter(listObj)
return data.most_common(1)
def timeIt(f, *args):
import time
start = time.clock()
m = f(*args)
return m, (time.clock() - start)*10
def findWords(count, Qdict = self.qd3g_f, depth = 3, lengthObj = self.length_f):
words = []
for i in range(count):
if type(lengthObj) == tuple:
length = self.gc.findLength(lengthObj)
else:
length = lengthObj
word = self.gc.findWord(Qdict, depth, length)
words.append(word)
return words
def compareMethods(afile, oc, qd3g, qd2g, prob):
probTotal1 = []
probTotal2 = []
probTotal3 = []
print("Mõõdiku võrdlus:")
for i in range(10000):
word1 = self.random_line(afile) #originaal
wordlen = len(word1)
word2 = oc.findWord(qd2g, 2, wordlen) #digraaf
word3 = oc.findWord(qd3g, 3, wordlen) #trigraaf
prob1 = oc.findWordProbability(word1,prob,1) #* 10**6
prob2 = oc.findWordProbability(word2,prob,1) #* 10**6
prob3 = oc.findWordProbability(word3,prob,1) #* 10**6
probTotal1.append(prob1)
probTotal2.append(prob2)
probTotal3.append(prob3)
#print ("Sõna1: %s Tõenäosus: %.3g" % (word1, prob1))
#print ("Sõna2: %s Tõenäosus: %.3g" % (word2, prob2))
#print ("Sõna3: %s Tõenäosus: %.3g" % (word3, prob3))
print('ORIGINAAL min, avg, median, max tõenäosus 1: %.3g, %.3g, %.3g, %.3g' % (min(probTotal1),avg(probTotal1), median(probTotal1),max(probTotal1)))
print('DIGRAMM min, avg, , median, max tõenäosus 2: %.3g, %.3g, %.3g, %.3g' % (min(probTotal2),avg(probTotal2), median(probTotal2), max(probTotal2)))
print('TRIGRAMM min, avg, , median, max tõenäosus 3: %.3g, %.3g, %.3g, %.3g' % (min(probTotal3),avg(probTotal3), median(probTotal3), max(probTotal3)))
def lengthTest(gc,length):
lengths = []
for i in range(100000):
lengths.append(gc.findLength(length))
return min(lengths), avg(lengths), median(lengths), max(lengths)
def findMetric(words, gc, probQD):
for word in words:
print ('Sõna: %s tõenäosus on: %.3g' % (word, gc.findWordProbability(word,probQD,1)))
def cgiOutput():
def chooseAndGet():
form = cgi.FieldStorage()
result = {}
len = None
if 'len' in form:
try:
if int(form['len'].value) < 50 and int(form['len'].value) > 2:
len = int(form['len'].value)
except ValueError:
pass
if 'how' in form:
if form['how'].value == 'dgf':
if not len:
len = self.length_f
result = findWords(20, self.qd2g_f, 2, len)
elif form['how'].value == 'tgf':
if not len:
len = self.length_f
result = findWords(20, self.qd3g_f, 3, len)
elif form['how'].value == 'dgh':
if not len:
len = self.length_h
result = findWords(20, self.qd2g_h, 2, len)
elif form['how'].value == 'tgh':
if not len:
len = self.length_h
result = findWords(20, self.qd3g_h, 3, len)
return json.dumps(result)
print ("Content-type:text/html;encoding=UTF-8")
print ()
if ('REQUEST_METHOD' in os.environ and
os.environ['REQUEST_METHOD'] == 'POST'):
print(chooseAndGet())
def shellOutput():
#print(lengthTest(self.gc, self.length_f))
#words = ('htamasvu', 'seekalti', 'hapäraus', 'lilestis', 'tisilmma')
words = ('eektrond', 'kuvkarkp', 'isatopun', 'apakseli', 'mergurmi',
'aksindik', 'kipeenam', 'põhjuhem', 'sidussaa', 'annstann')
words = ('kaigenda', 'päikulte', 'liidusam', 'metastap', 'väivolis',
'väivolis', 'palbapre', 'soongiaa', 'ereelatu', 'nargaste', 'tsülduku')
#words = ('tiiminst', 'almilinn', 'rotselva', 'vahulinv', 'noomumbi')
#words = ('haigeduv', 'sistamat', 'igatsivi', 'särklaag', 'andrilin')
#words = ('htamasvu','õhesteni', 'oorusaja', 'tiuspial', 'avõhuria', 'tasimisu')
#findMetric(words, self.gc, self.qd1g_f)
'''
a = compareMethods('resource/lemmad_utf.txt',
self.gc, self.qd3g_h,
self.qd2g_h, self.qd1g_h)
'''
'''
lst = []
with open('resource/lemmad_utf.txt', 'r') as f:
for l in f.readlines():
lst.append(len(l))
print ('Min: %s, Avg: %s Keskv: %s Max: %s' % (min(lst), avg(lst), median(lst),max(lst)))
'''
if 'GATEWAY_INTERFACE' in os.environ:
cgiOutput()
else:
shellOutput()
return
def save(self):
def scanAndSave():
qd = GraphCollection()
fileName1 = 'resource/lemmad_utf.txt'
fileName2 = 'resource/lemmad_piir_utf.txt'
trie1 = qd.createTrie(fileName1, 3)
trie2 = qd.createTrie(fileName2, 3)
length1 = qd.createLengthQuickArr(fileName1)
length2 = qd.createLengthQuickArr(fileName2)
probability1 = qd.createQuickDict(trie1, 1)
probability2 = qd.createQuickDict(trie2, 1)
generatorD1 = qd.createQuickDict(trie1, 2)
generatorD2 = qd.createQuickDict(trie2, 2)
generatorT1 = qd.createQuickDict(trie1, 3)
generatorT2 = qd.createQuickDict(trie2, 3)
qd.save('resource/qd3g_f.pck', generatorT1)
qd.save('resource/qd2g_f.pck', generatorD1)
qd.save('resource/qd1g_f.pck', probability1)
qd.save('resource/lngt_f.pck', length1)
qd.save('resource/qd3g_h.pck', generatorT2)
qd.save('resource/qd2g_h.pck', generatorD2)
qd.save('resource/qd1g_h.pck', probability2)
qd.save('resource/lngt_h.pck', length2)
scanAndSave()
def load(self):
def loadOptimized(self):
self.gc = GraphCollection()
self.qd1g_f = self.gc.load('resource/qd1g_f.pck')
self.qd2g_f = self.gc.load('resource/qd2g_f.pck')
self.qd3g_f = self.gc.load('resource/qd3g_f.pck')
self.length_f = self.gc.load('resource/lngt_f.pck')
self.qd1g_h = self.gc.load('resource/qd1g_h.pck')
self.qd2g_h = self.gc.load('resource/qd2g_h.pck')
self.qd3g_h = self.gc.load('resource/qd3g_h.pck')
self.length_h = self.gc.load('resource/lngt_h.pck')
loadOptimized(self)
def random_line(self,afile):
return random.choice(list(open(afile))).rstrip('\n')
