def filter(ham_word_pro, spam_word_pro, test_file):
test_paths = fun(test_file)
for test_path in test_paths:
email_spam_prob = 0.0
spam_prob = 0.5
ham_prob = 0.5
file_name = test_path.split('\\')[-1]
prob_dict = {}
words = set(email_parser(test_path))
for word in words:
Psw = 0.0
if word not in spam_word_pro:
Psw = 0.4
else:
Pws = spam_word_pro[word]
Pwh = ham_word_pro[word]
Psw = spam_prob*(Pws/(Pwh*ham_prob+Pws*spam_prob))
prob_dict[word] = Psw
numerator = 1
denominator_h = 1
for k, v in prob_dict.items():
numerator *= v
denominator_h *= (1-v)
email_spam_prob = round(numerator/(numerator+denominator_h), 4)
if email_spam_prob > 0.5:
print(file_name, 'spam', email_spam_prob)
else:
print(file_name, 'ham', email_spam_prob)
print(prob_dict)
print('******************************************************')
def main():
file_path = r'..\data\data_of_movie'
# file_path = r'..\data\test'
output_path = r'..\data\tenTimesTraining'
files = fun(file_path)
# output_path = buildfile(output_path)
data_storage = split_ten(files)
group_data(data_storage, output_path)
def get_word(email_file):
word_list = []
word_set = []
punctuations = """,.<>()*&^%$#@!'";~`[]{}|、\\/~+_-=?"""
email_paths = fun(email_file)
for email_path in email_paths:
clean_word = email_parser(email_path)
word_list.append(clean_word)
word_set.extend(clean_word)
return word_list, set(word_set)
def get_data(data_path):
label_vec = []
files = fwalker.fun(data_path)
for file in files:
ech_label_vec = []
ech_label = int((file.split('\\'))[-1][0])
ech_vec = ((np.loadtxt(file)).ravel())
ech_label_vec.append(ech_label)
ech_label_vec.append(ech_vec)
label_vec.append(ech_label_vec)
return label_vec
def main():
filepath = r'..\email'
testpath = r'..\test'
files = fun(filepath)
random.shuffle(files)
top10 = files[:10]
for ech in top10:
ech_name = testpath + '\\' + ('_'.join(ech.split('\\')[-2:]))
shutil.move(ech, testpath)
os.rename(testpath + '\\' + ech.split('\\')[-1], ech_name)
print('%s moved' % ech_name)
def get_word(email_file):
word_list = []
word_set = []
punctuations = """,.<>()*&^%$#@!'";~`[]{}|、\\/~+_-=?"""
email_paths = fun(email_file)
for email_path in email_paths:
# content_list = readtxt(email_path, 'utf8')
# content = (' '.join(content_list)).replace(
# '\r\n', ' ').replace('\t', ' ')
# for punctuation in punctuations:
# # content = content.replace(punctuation, '').replace(' ', ' ')
# content = (' '.join(content.split(punctuation))).replace(' ', ' ')
# clean_word = [word.lower()
# for word in content.split(' ') if len(word) > 2]
clean_word = email_parser(email_path)
word_list.append(clean_word)
word_set.extend(clean_word)
return word_list, set(word_set)
def main():
datapath = r'D:\DevelopmentLanguage\Python\MachineLearning\KNN\lab3_0930\digits'
inputpath = bfile.buildfile(
r'D:\DevelopmentLanguage\Python\MachineLearning\KNN\lab3_0930\input_digits')
files = fwalker.fun(datapath)
change_data(files, inputpath)