update steps

src/main.py

@@ -29,14 +29,14 @@ def main():
     device_df, diary_df, click_df, conversion_df = diary_fe.read_csv_data(Path("/srv/apps/node2vec_git/cvr_data/"))
     # print(diary_df.sample(1))
     device_df = device_fe.device_feature_engineering(device_df)
-    print(device_df.sample(1))
+    # print(device_df.sample(1))
     diary_df = diary_fe.diary_feature_engineering(diary_df)
-    print(diary_df.sample(1))
+    # print(diary_df.sample(1))
     cc_df = diary_fe.click_feature_engineering(click_df, conversion_df)
-    print(cc_df.sample(1))
+    # print(cc_df.sample(1))
     df = diary_fe.join_features(device_df, diary_df, cc_df)
-    print(df.sample(1))
-    print(df.dtypes)
+    # print(df.sample(1))
+    # print(df.dtypes)
 
     train_df, test_df = train_test_split(df, test_size=0.2)
     train_df, val_df = train_test_split(train_df, test_size=0.2)
@@ -53,8 +53,7 @@ def main():
     estimator_config = tf.estimator.RunConfig(session_config=session_config)
 
     model = tf.estimator.Estimator(model_fn=esmm_model_fn, params=params, model_dir=model_path, config=estimator_config)
-    # TODO 50000
-    train_spec = tf.estimator.TrainSpec(input_fn=lambda: esmm_input_fn(train_df, shuffle=True), max_steps=20000)
+    train_spec = tf.estimator.TrainSpec(input_fn=lambda: esmm_input_fn(train_df, shuffle=True), max_steps=50000)
     eval_spec = tf.estimator.EvalSpec(input_fn=lambda: esmm_input_fn(val_df, shuffle=False))
     tf.estimator.train_and_evaluate(model, train_spec, eval_spec)
 
@@ -74,7 +73,7 @@ def main():
 
     predict_fn = tf.contrib.predictor.from_saved_model(save_path)
 
-    print("==============================")
+    print("============================================================")
     # device_id = "861601036552944"
     # diary_ids = [
     #     "16195283", "16838351", "17161073", "17297878", "17307484", "17396235", "16418737", "16995481", "17312201", "12237988"

src/models/esmm/input_fn.py

@@ -5,6 +5,7 @@ from .utils import create_boundaries, create_vocabulary_list
 
 
 def build_features(df):
+    # TODO
     int_columns = ["active_days", "topic_num", "favor_num", "vote_num"]
     float_columns = ["one_ctr", "three_ctr", "seven_ctr", "fifteen_ctr"]
     numeric_features = []
@@ -14,6 +15,7 @@ def build_features(df):
         else:
             numeric_features.append(fc.bucketized_column(fc.numeric_column(col), boundaries=create_boundaries(df, col)))
 
+    # TODO
     categorical_columns = [
         "device_id", "active_type", "past_consume_ability_history", "potential_consume_ability_history",
         "price_sensitive_history", "card_id", "is_pure_author", "is_have_reply", "is_have_pure_reply", "content_level",

src/models/esmm/model.py

@@ -99,6 +99,7 @@ def model_predict_diary(device_id, diary_ids, device_dict, diary_dict, predict_f
         time_1 = timeit.default_timer()
         device_info, diary_lst, diary_ids_res = device_diary_fe(device_id, diary_ids, device_dict, diary_dict)
         print("predict check: " + str(len(diary_lst)) + " " + str(len(diary_ids_res)))
+        # TODO
         int_columns = [
             "active_type", "active_days", "card_id", "is_pure_author", "is_have_reply", "is_have_pure_reply", "content_level",
             "topic_num", "favor_num", "vote_num"