lalithadevi
commited on
Commit
•
a364621
1
Parent(s):
367d701
Update news_category_similar_news_prediction.py
Browse files
news_category_similar_news_prediction.py
CHANGED
@@ -170,12 +170,6 @@ def predict_news_category_similar_news(old_news: pd.DataFrame, new_news: pd.Data
|
|
170 |
final_df['category'] = label
|
171 |
final_df['pred_proba'] = prob
|
172 |
final_df['similar_news'] = sim_news
|
173 |
-
# final_df.reset_index(drop=True, inplace=True)
|
174 |
-
# final_df.loc[final_df['pred_proba']<CLASSIFIER_THRESHOLD, 'category'] = 'NATION'
|
175 |
-
# final_df.loc[(final_df['title'].str.contains('Pakistan')) & (final_df['category'] == 'NATION'), 'category'] = 'WORLD'
|
176 |
-
# logger.warning('Updated category of articles having Pakistan in title and category=NATION to WORLD')
|
177 |
-
# final_df.loc[(final_df['title'].str.contains('Zodiac Sign', case=False)) | (final_df['title'].str.contains('Horoscope', case=False)), 'category'] = 'SCIENCE'
|
178 |
-
# logger.warning('Updated category of articles having Zodiac Sign in title to SCIENCE')
|
179 |
|
180 |
final_df = process_prediction_df(final_df, df_type="production & prediction")
|
181 |
prediction_df = final_df.copy()
|
@@ -208,13 +202,6 @@ def predict_news_category_similar_news(old_news: pd.DataFrame, new_news: pd.Data
|
|
208 |
prediction_df = new_news.copy()
|
209 |
prediction_df = add_flags_to_prediction_df(prediction_df)
|
210 |
|
211 |
-
# final_df.drop_duplicates(subset='url', keep='first', inplace=True)
|
212 |
-
# final_df.reset_index(drop=True, inplace=True)
|
213 |
-
# final_df.loc[final_df['pred_proba']<CLASSIFIER_THRESHOLD, 'category'] = 'NATION'
|
214 |
-
# final_df.loc[(final_df['title'].str.contains('Pakistan')) & (final_df['category'] == 'NATION'), 'category'] = 'WORLD'
|
215 |
-
# logger.warning('Updated category of articles having Pakistan in title and category=NATION to WORLD')
|
216 |
-
# final_df.loc[(final_df['title'].str.contains('Zodiac Sign', case=False)) | (final_df['title'].str.contains('Horoscope', case=False)), 'category'] = 'SCIENCE'
|
217 |
-
# logger.warning('Updated category of articles having Zodiac Sign in title to SCIENCE')
|
218 |
else:
|
219 |
logger.warning('INFO: Old & New Articles are the same. There is no requirement of updating them in the database. Database is not updated.')
|
220 |
db_updation_required = 0
|
|
|
170 |
final_df['category'] = label
|
171 |
final_df['pred_proba'] = prob
|
172 |
final_df['similar_news'] = sim_news
|
|
|
|
|
|
|
|
|
|
|
|
|
173 |
|
174 |
final_df = process_prediction_df(final_df, df_type="production & prediction")
|
175 |
prediction_df = final_df.copy()
|
|
|
202 |
prediction_df = new_news.copy()
|
203 |
prediction_df = add_flags_to_prediction_df(prediction_df)
|
204 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
205 |
else:
|
206 |
logger.warning('INFO: Old & New Articles are the same. There is no requirement of updating them in the database. Database is not updated.')
|
207 |
db_updation_required = 0
|