get print data

word_vector/answer.py

@@ -6,12 +6,16 @@ from collections import defaultdict
 
 sys.path.append(os.path.realpath("."))
 
+import random
+
 from gensim.models import Word2Vec, word2vec
-from utils.date import get_ndays_before_no_minus, get_ndays_before_with_format
+from utils.date import (get_ndays_before_no_minus,
+    get_ndays_before_with_format)
 from utils.es import get_online_ids
 from utils.files import DATA_PATH, MODEL_PATH
 from utils.spark import get_spark
 
+
 answer_click_ids_model_path = os.path.join(MODEL_PATH, "answer_click_ids_item2vec_model")
 
 try:
@@ -31,7 +35,7 @@ def get_answer_click_data(spark, start, end):
           where action = 'page_view'
             AND partition_date BETWEEN '{}' AND '{}'
             AND page_name='answer_detail'
-            AND page_stay>=1
+            AND page_stay>=0.5
             AND cl_id is not null
             AND cl_id != ''
             AND business_id is not null
@@ -179,6 +183,7 @@ if __name__ == "__main__":
     print(click_df.count())
 
     res_dict = get_device_click_answer_ids_dict(click_df)
+    print(random.sample(res_dict.items(), 3))
 
     with open(os.path.join(DATA_PATH, "click_answer_ids.csv"), "w") as f:
         for (k, v) in res_dict.items():