def demo4():
from nltk_contrib import classify
from nltk import detect
from nltk.corpora import genesis
from itertools import islice
fd = detect.feature({
"2-tup":
lambda t: [' '.join(t)[n:n + 2] for n in range(len(' '.join(t)) - 1)],
"words":
lambda t: t
})
classifier = classify.NaiveBayes(fd)
training_data = {}
training_data["english-kjv"] = list(
islice(genesis.raw("english-kjv"), 0, 400))
training_data["french"] = list(islice(genesis.raw("french"), 0, 400))
training_data["finnish"] = list(islice(genesis.raw("finnish"), 0, 400))
classifier.train(training_data)
result = classifier.get_class_probs(
list(islice(genesis.raw("english-kjv"), 150, 200)))
print 'english-kjv :', result.prob('english-kjv')
print 'french :', result.prob('french')
print 'finnish :', result.prob('finnish')
def demo():
from nltk.corpus import brown
from nltk import detect
detector = detect.feature({"initial": lambda t: [t[0]], "len": lambda t: [len(t)]})
for sent in brown.words("a")[:10]:
print detector(sent)
def demo():
from nltk.corpus import brown
from nltk import detect
detector = detect.feature({
'initial': lambda t: [t[0]],
'len': lambda t: [len(t)]
})
for sent in brown.words('a')[:10]:
print detector(sent)
def demo():
from nltk_contrib import classify
from nltk import detect
fd = detect.feature({"1-tup": lambda t: list(t)})
classifier = classify.NaiveBayes(fd)
training_data = {"class a": "aaaaaab", "class b": "bbbbbba"}
classifier.train(training_data)
result = classifier.get_class_dict("a")
for cls in result:
print cls, ':', result[cls]
"""
def demo():
from nltk_contrib import classify
from nltk import detect
fd = detect.feature({"1-tup": lambda t: [t[n] for n in range(len(t))]})
classifier = classify.spearman.Spearman(fd)
trainning_data = {"class a": "aaaaaab", "class b": "bbbbbba"}
classifier.train(trainning_data)
result = classifier.get_class_dict("a")
for cls in result:
print cls, ':', result[cls]
"""
def demo2():
from nltk_contrib import classify
from nltk import detect
fd = detect.feature(
{"2-tup": lambda t: [t[n:n + 2] for n in range(len(t))]})
classifier = classify.NaiveBayes(fd)
training_data = {"class a": "aaaaaab", "class b": "bbbbbba"}
classifier.train(training_data)
result = classifier.get_class_dict("aababb")
for cls in result:
print cls, ':', result[cls]
"""
def demo2():
from nltk_contrib import classify
from nltk import detect
fd = detect.feature({"2-tup": lambda t: [t[n:n+2] for n in range(len(t))]})
classifier = classify.NaiveBayes(fd)
training_data = {"class a": "aaaaaab",
"class b": "bbbbbba"}
classifier.train(training_data)
result = classifier.get_class_dict("aababb")
for cls in result:
print cls, ':', result[cls]
"""
def demo():
from nltk_contrib import classify
from nltk import detect
fd = detect.feature({"1-tup": lambda t: [t[n] for n in range(len(t))]})
classifier = classify.spearman.Spearman(fd)
trainning_data = {"class a": "aaaaaab",
"class b": "bbbbbba"}
classifier.train(trainning_data)
result = classifier.get_class_dict("a")
for cls in result:
print cls, ':', result[cls]
"""
def demo():
from nltk_contrib import classify
from nltk import detect
fd = detect.feature({"1-tup": lambda t: list(t)})
classifier = classify.NaiveBayes(fd)
training_data = {"class a": "aaaaaab",
"class b": "bbbbbba"}
classifier.train(training_data)
result = classifier.get_class_dict("a")
for cls in result:
print cls, ':', result[cls]
"""
def demo4():
from nltk_contrib import classify
from nltk import detect
from nltk.corpora import genesis
from itertools import islice
fd = detect.feature({"2-tup": lambda t: [' '.join(t)[n:n+2] for n in range(len(' '.join(t))-1)],
"words": lambda t: t})
classifier = classify.NaiveBayes(fd)
training_data = {}
training_data["english-kjv"] = list(islice(genesis.raw("english-kjv"), 0, 400))
training_data["french"] = list(islice(genesis.raw("french"), 0, 400))
training_data["finnish"] = list(islice(genesis.raw("finnish"), 0, 400))
classifier.train(training_data)
result = classifier.get_class_probs(list(islice(genesis.raw("english-kjv"), 150, 200)))
print 'english-kjv :', result.prob('english-kjv')
print 'french :', result.prob('french')
print 'finnish :', result.prob('finnish')
from nltk_contrib import classify
from nltk import detect
from nltk.corpus import udhr
import string
def run(classifier, training_data, gold_data):
classifier.train(training_data)
correct = 0
for lang in gold_data:
cls = classifier.get_class(gold_data[lang])
if cls == lang:
correct += 1
print correct, "in", len(gold_data), "correct"
# features: character bigrams
fd = detect.feature({"char-bigrams" : lambda t: [string.join(t)[n:n+2] for n in range(len(t)-1)]})
training_data = udhr.langs(['English-Latin1', 'French_Francais-Latin1', 'Indonesian-Latin1', 'Zapoteco-Latin1'])
gold_data = {}
for lang in training_data:
gold_data[lang] = training_data[lang][:50]
training_data[lang] = training_data[lang][100:200]
print "Cosine classifier: ",
run(classify.Cosine(fd), training_data, gold_data)
print "Naivebayes classifier: ",
run(classify.NaiveBayes(fd), training_data, gold_data)
print "Spearman classifier: ",
run(classify.Spearman(fd), training_data, gold_data)
import string
def run(classifier, training_data, gold_data):
classifier.train(training_data)
correct = 0
for lang in gold_data:
cls = classifier.get_class(gold_data[lang])
if cls == lang:
correct += 1
print correct, "in", len(gold_data), "correct"
# features: character bigrams
fd = detect.feature({
"char-bigrams":
lambda t: [string.join(t)[n:n + 2] for n in range(len(t) - 1)]
})
training_data = udhr.langs([
'English-Latin1', 'French_Francais-Latin1', 'Indonesian-Latin1',
'Zapoteco-Latin1'
])
gold_data = {}
for lang in training_data:
gold_data[lang] = training_data[lang][:50]
training_data[lang] = training_data[lang][100:200]
print "Cosine classifier: ",
run(classify.Cosine(fd), training_data, gold_data)
print "Naivebayes classifier: ",
