update field

src/main.py

@@ -19,20 +19,18 @@ from models.esmm.model import esmm_model_fn, model_export, model_predict
 def main():
     time_begin = time.time()
     # df = get_device_df_from_redis()
-    df2 = get_diary_df_from_redis()
-    print(df2.sample(1))
-    print(df2.columns)
+    # df2 = get_diary_df_from_redis()
+    # print(df2.sample(1))
     # print(df.size)
     # print(df2.size)
 
     # a = device_feature_engineering(df)
-    b = diary_feature_engineering(df2)
     # print(a.size)
-    print(b.sample(1))
+    # b = diary_feature_engineering(df2)
+    # print(b.sample(1))
 
     device_df, diary_df, click_df, conversion_df = read_csv_data(Path("~/data/cvr_data/"))
     print(diary_df.sample(1))
-    print(diary_df.columns)
     # device_df = device_feature_engineering(device_df)
     diary_df = diary_feature_engineering(diary_df)
     print(diary_df.sample(1))

src/models/esmm/fe.py

@@ -86,9 +86,9 @@ def diary_feature_engineering(df):
     diary_df["second_positions"] = diary_df["second_positions"].apply(lambda d: d if isinstance(d, list) else [])
     diary_df["projects"] = diary_df["projects"].apply(lambda d: d if isinstance(d, list) else [])
 
-    diary_df["is_pure_author"] = diary_df["is_pure_author"].map(str_bool_map)
-    diary_df["is_have_pure_reply"] = diary_df["is_have_pure_reply"].map(str_bool_map)
-    diary_df["is_have_reply"] = diary_df["is_have_reply"].map(str_bool_map)
+    # diary_df["is_pure_author"] = diary_df["is_pure_author"].map(str_bool_map)
+    # diary_df["is_have_pure_reply"] = diary_df["is_have_pure_reply"].map(str_bool_map)
+    # diary_df["is_have_reply"] = diary_df["is_have_reply"].map(str_bool_map)
 
     diary_df["is_pure_author"] = diary_df["is_pure_author"].astype(int)
     diary_df["is_have_pure_reply"] = diary_df["is_have_pure_reply"].astype(int)