def sort_article_list(select_date):
'''функция сотировки по дате
'''
articles = db.get_articles()
sorted_articles = []
for article in articles:
if article['data_create'] == select_date:
sorted_articles.append(article)
update_article_list(sorted_articles)
def request_user():
'''получаем ответ от пользователя на удаления данных
'''
articles = db.get_articles()
answer = messagebox.askyesno(title="Вопрос", message="Удалить данные?")
if answer == True:
articles.clear()
db.write_articles(articles)
demo()
def result_income():
'''считаем наши все доходы
'''
income_list = []
articles = db.get_articles()
for article in articles:
if article['type_item'] == '+':
income_list.append(article['cost'])
summ_income = sum(income_list)
return summ_income
def result_expence():
'''считаем все наши расходы
'''
expence_list = []
articles = db.get_articles()
for article in articles:
if article['type_item'] == '-':
expence_list.append(article['cost'])
summ_expence = sum(expence_list)
return summ_expence
def del_article():
'''удалаляем в окне не нужный
пунк дохода или расхода.
'''
articles = db.get_articles()
select = list(text_result.get())
select.reverse()
for i in select:
text_result.delete(i)
articles.pop(i)
db.write_articles(articles)
def sort_article_range_date(start_date, end_date):
'''функция определения дат по заданным периодам
'''
articles = db.get_articles()
sorted_articles = []
date_1 = min(start_date, end_date)
date_2 = max(start_date, end_date)
while date_1 < date_2:
date_1 += datetime.timedelta(days=1)
for article in articles:
if article['data_create'] == date_1:
sorted_articles.append(article)
update_article_list(sorted_articles)
def update_article_list():
'''выводим наши все записи
в окно
'''
text_result.delete(0, END)
index = 0
articles = db.get_articles()
for article in articles:
item_value = article['type_value']
s = '{cost} руб. {type_item} {type_value} {data_create}'.format(
get_short_string(item_value), **article)
text_result.insert(0, s)
if article['type_item'] == '-':
text_result.itemconfig(index, bg='red')
else:
text_result.itemconfig(index, bg='green')
index += 1
def add_type_rashod():
'''добавляем в спикок словарь с нашими расходами
'''
articles = db.get_articles()
#перехватываем ошибку неправильного ввода
try:
result_cost = int(ent_rashod_dohod.get())
except ValueError as err:
mb.showerror('Ошибка', 'Должно быть введено число!')
new_article = {
'cost': result_cost,
'type_item': '-',
'type_value': variable.get(),
'data_create': datetime.datetime.today()
}
#widgets.ent_rashod_dohod.delete(0, END)
articles.append(new_article)
db.write_articles(articles)
update_article_list()
import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer, TfidfTransformer
from sklearn.pipeline import Pipeline
from nltk.stem.snowball import SnowballStemmer
from scipy.sparse import hstack
from sklearn.metrics import confusion_matrix, classification_report, auc, roc_curve
from sklearn.model_selection import cross_val_score
import matplotlib.pyplot as plt
import itertools
import pandas as pd
import rethinkdb as r
import re
articles = get_articles(lambda join: {
'body': join['left']['body'],
'title': join['left']['title'],
'credible': join['right']['isReliable']
})
df = pd.DataFrame.from_records(articles)
with open('./tl_stopwords.txt', 'r') as f:
TL_STOPWORDS = f.read().splitlines()
STOP_WORDS = ENGLISH_STOP_WORDS.union(TL_STOPWORDS)
body_tfidf = TfidfVectorizer(
token_pattern=r'(?ui)\b\w*[a-z]{2}\w*\b',
stop_words=STOP_WORDS,
ngram_range=(1, 2),
max_df=0.85,
plt.ylabel('True label')
plt.xlabel('Predicted label')
articles = get_articles(
lambda join: {
'body': join['left']['body'],
'title': join['left']['title'],
# 'src_social_score': join['right']['socialScore'],
# 'src_has_impt_pages': r.branch(
# join['right']['contactUsUrl'],
# 1,
# join['right']['aboutUsUrl'],
# 1,
# 0
# ),
'src_has_about': 1 if join['right']['aboutUsUrl'] else 0,
'src_has_contact': 1 if join['right']['contactUsUrl'] else 0,
# 'src_wot_reputation': join['right']['wotReputation'],
# 'src_country_rank': join['right']['countryRank'],
'src_world_rank': join['right']['worldRank'],
'src_domain_has_number': 1 if join['right']['domainHasNumber'] else 0,
'src_domain_is_blog': 1 if join['right']['isBlogDomain'] else 0,
# 'src_domain_is_suspicious': 1 if join['right']['isDomainSuspicious'] else 0,
# 'src_domain_creation_date': join['right']['domainCreationDate'],
'src_id': join['right']['id'],
'credible': join['right']['isReliable']
})
train_sources = get_sources_sample(20)
train_articles = [a for a in articles if a['src_id'] not in train_sources]