Merge branch 'fe' into 'offic'

Fe

See merge request !8

src/models/esmm/fe/click_fe.py

@@ -6,7 +6,7 @@ def click_feature_engineering(click_df, conversion_df):
     # conversion_df = conversion_df.copy()
 
     click_df.rename(columns={"label": "click_label"}, inplace=True)
-    click_df = click_df.sample(conversion_df.shape[0] * 10)
+    # click_df = click_df.sample(conversion_df.shape[0] * 10)
     print("click_df: " + str(click_df.shape))
     conversion_df.rename(columns={"label": "conversion_label"}, inplace=True)
     print("conversion_df: " + str(conversion_df.shape))

src/online_prediction.py

@@ -20,7 +20,7 @@ def consume_kafka(device_dict, diary_dict, tractate_dict, diary_predict_fn, trac
     consumer.poll()
     consumer.seek_to_end()
 
-    predict_size = 300
+    predict_size = 200
 
     try:
         for msg in consumer:

src/train_diary.py

@@ -50,7 +50,7 @@ def main():
     train_df, val_df = train_test_split(train_df, test_size=0.2)
 
     all_features = fe.build_features(df, diary_fe.INT_COLUMNS, diary_fe.FLOAT_COLUMNS, diary_fe.CATEGORICAL_COLUMNS)
-    params = {"feature_columns": all_features, "hidden_units": [64, 32], "learning_rate": 0.1}
+    params = {"feature_columns": all_features, "hidden_units": [360, 200, 80, 2], "learning_rate": 0.1}
     model_path = str(Path("/data/files/model_tmp/diary/").expanduser())
     if os.path.exists(model_path):
         shutil.rmtree(model_path)

src/train_tractate.py

@@ -47,7 +47,7 @@ def main():
     train_df, val_df = train_test_split(train_df, test_size=0.2)
 
     all_features = fe.build_features(df, tractate_fe.INT_COLUMNS, tractate_fe.FLOAT_COLUMNS, tractate_fe.CATEGORICAL_COLUMNS)
-    params = {"feature_columns": all_features, "hidden_units": [128, 64, 32], "learning_rate": 0.1}
+    params = {"feature_columns": all_features, "hidden_units": [360, 200, 80, 2], "learning_rate": 0.2}
     model_path = str(Path("/data/files/model_tmp/tractate/").expanduser())
     if os.path.exists(model_path):
         shutil.rmtree(model_path)