try read

src/main_portrait.py

@@ -59,10 +59,14 @@ def main():
     save_path = "/home/gmuser/data/models/1595501144"
     predict_fn = tf.contrib.predictor.from_saved_model(save_path)
 
-    # device_id = "androidid_a25a1129c0b38f7b"
+    device_id = "androidid_a25a1129c0b38f7b"
     # offline_predict(device_id, device_dict, diary_dict, predict_fn)
 
-    user_portrait_scan_info(device_dict, diary_dict, predict_fn)
+    res = user_portrait_tag3_get_candidate_unread_list(device_id, "diary")
+    print(len(res))
+    print(res[:10])
+
+    # user_portrait_scan_info(device_dict, diary_dict, predict_fn)
 
 
 if __name__ == "__main__":

src/utils/portrait.py

@@ -69,6 +69,20 @@ def remove_read_with_size(lol, content_type, read_list, size=500):
     return res
 
 
+def get_n_unread_data_from_ids(data, size, read_lst):
+    """
+    data: ['852170', '864084', '231589']
+    """
+    res = []
+    for id_str in data:
+        id = int(id_str)
+        if len(res) >= size:
+            break
+        elif (id not in set(read_lst)) and (id not in res):
+            res.append(id)
+    return res
+
+
 def _user_portrait_tag3_get_ctcvr_candidate_key(device_id, content_type):
     """
     device_id:
@@ -128,8 +142,10 @@ def user_portrait_tag3_get_candidate_unread_list(device_id, content_type, size=5
     read_list, _ = get_user_portrait_tag3_read_v2(device_id, content_type)
     diary_dict = user_portrait_tag3_get_candidate_dict(device_id, "diary")
     if diary_dict:
-        lol = remove_read_with_size(diary_dict.values(), "diary", read_list, size)
+        # lol = remove_read_with_size(diary_dict.values(), "diary", read_list, size)
+        lol = diary_dict.values()
         queue = merge_lists(*lol)
+        get_n_unread_data_from_ids(queue, size, read_list)
     return queue