update path

word_vector/api.py

@@ -12,7 +12,7 @@ from utils.es import es_scan
 from utils.files import DATA_PATH, MODEL_PATH
 from utils.message import send_msg_to_dingtalk
 
-from word_vector.tractate import tractate_click_ids_model_path
+from word_vector.tractate import TRACTATE_CLICK_IDS_MODEL
 
 model_output_name = "w2v_model"
 model_path = os.path.join(MODEL_PATH, model_output_name)
@@ -21,11 +21,6 @@ try:
 except Exception as e:
     print(e)
 
-try:
-    TRACTATE_CLICK_IDS_MODEL = word2vec.Word2Vec.load(tractate_click_ids_model_path)
-except Exception as e:
-    print(e)
-
 
 class W2vSentences:
     def __init__(self, f_name):

word_vector/tractate.py

 import multiprocessing
 import os
 import sys
+import time
 from collections import defaultdict
 
 sys.path.append(os.path.realpath("."))
 
-import time
-
 from gensim.models import Word2Vec, word2vec
 from utils.date import get_ndays_before_no_minus, get_ndays_before_with_format
 from utils.files import DATA_PATH, MODEL_PATH
 from utils.spark import get_spark
 
-from word_vector.api import clicked_tractate_ids_item2vec_model
-
 tractate_click_ids_model_path = os.path.join(MODEL_PATH, "tractate_click_ids_item2vec_model")
 
+try:
+    TRACTATE_CLICK_IDS_MODEL = word2vec.Word2Vec.load(tractate_click_ids_model_path)
+except Exception as e:
+    print(e)
+
 
 def get_tracate_click_data(spark, start, end):
     reg = r"""^\\d+$"""
@@ -165,7 +167,7 @@ if __name__ == "__main__":
     save_clicked_tractate_ids_item2vec()
 
     for id in ["84375", "148764", "368399"]:
-        print(clicked_tractate_ids_item2vec_model(id, n=5))
+        print(TRACTATE_CLICK_IDS_MODEL.wv.most_similar(id, topn=5))
 
     print("total cost: {:.2f}mins".format((time.time() - begin_time) / 60))