update

region_rule_rank_h.py

@@ -4,6 +4,7 @@
 # @Time: 2022/5/5 15:54
 # @Software: PyCharm
 
+import os
 import gevent
 import datetime
 import pandas as pd
@@ -16,6 +17,8 @@ from config import set_config
 from log import Log
 from check_video_limit_distribute import update_limit_video_score
 
+os.environ['NUMEXPR_MAX_THREADS'] = '16'
+
 config_, _ = set_config()
 log_ = Log()
 
@@ -450,6 +453,7 @@ def rank_by_h(project, table, now_date, now_h, rule_params, region_code_list):
             rule_key = param.get('rule')
             rule_param = rule_params_item.get(rule_key)
             log_.info(f"rule_key = {rule_key}, rule_param = {rule_param}")
+
             task_list = []
             for region in region_code_list:
                 t = Thread(target=process_with_region,

region_rule_rank_h_by24h.py

@@ -4,6 +4,8 @@
 # @Time: 2022/5/5 15:54
 # @Software: PyCharm
 
+import time
+import os
 import gevent
 import datetime
 import pandas as pd
@@ -15,6 +17,8 @@ from utils import RedisHelper, get_data_from_odps, filter_video_status, check_ta
 from config import set_config
 from log import Log
 
+os.environ['NUMEXPR_MAX_THREADS'] = '16'
+
 config_, _ = set_config()
 log_ = Log()
 
@@ -251,6 +255,13 @@ def rank_by_24h(project, table, now_date, now_h, rule_params, region_code_list):
             log_.info(f"rule_key = {rule_key}, rule_param = {rule_param}")
             df_list = [feature_df[feature_df['apptype'] == apptype] for apptype in data_param]
             df_merged = reduce(merge_df, df_list)
+
+            # thread_pool = ThreadPoolExecutor(max_workers=5)
+            # for region in region_code_list:
+            #     t = thread_pool.submit(process_with_region,
+            #                        (region, df_merged, app_type, data_key, rule_key, rule_param, now_date, now_h))
+            # thread_pool.shutdown(wait=True)
+
             task_list = []
             for region in region_code_list:
                 t = Thread(target=process_with_region,