microsoft · you-n-g · May 17, 2021 · Feb 16, 2021 · Feb 17, 2021 · Feb 26, 2021
diff --git a/qlib/__init__.py b/qlib/__init__.py
@@ -154,7 +154,7 @@ def init_from_yaml_conf(conf_path, **kwargs):
     init(default_conf, **config)
 
 
-def get_project_path(config_name="config.yaml") -> Path:
+def get_project_path(config_name="config.yaml", cur_path=None) -> Path:
     """
     If users are building a project follow the following pattern.
     - Qlib is a sub folder in project path
@@ -181,7 +181,8 @@ def get_project_path(config_name="config.yaml") -> Path:
     FileNotFoundError:
         If project path is not found
     """
-    cur_path = Path(__file__).absolute().resolve()
+    if cur_path is None:
+        cur_path = Path(__file__).absolute().resolve()
     while True:
         if (cur_path / config_name).exists():
             return cur_path
@@ -199,7 +200,7 @@ def auto_init(**kwargs):
     """
 
     try:
-        pp = get_project_path()
+        pp = get_project_path(cur_path=kwargs.pop("cur_path", None))
     except FileNotFoundError:
         init(**kwargs)
     else:

diff --git a/qlib/data/dataset/handler.py b/qlib/data/dataset/handler.py
@@ -50,6 +50,9 @@ class DataHandler(Serializable):
                    SH600004    13.313329  11800983.0       13.313329        13.317701    0.183632  0.0042
                    SH600005    37.796539  12231662.0       38.258602        37.919757    0.970325  0.0289
 
+
+    Tips for improving the performance of datahandler
+    - Fetching data with `col_set=CS_RAW` will return the raw data and may avoid pandas from copying the data when calling `loc`
     """
 
     def __init__(
@@ -257,6 +260,10 @@ def get_range_iterator(
 class DataHandlerLP(DataHandler):
     """
     DataHandler with **(L)earnable (P)rocessor**
+
+    Tips to improving the performance of data handler
+    - To reduce the memory cost
+        - `drop_raw=True`: this will modify the data inplace on raw data;
     """
 
     # data key

diff --git a/qlib/workflow/task/update.py b/qlib/workflow/task/update.py
@@ -97,8 +97,8 @@ def update_all_pred(self, rec_filter_func=None):
 
             .. code-block:: python
 
-                def rec_filter_func(recorder):
-                    task_config = recorder.load_object("task")
+                def record_filter(record):
+                    task_config = record.load_object("task")
                     if task_config["model"]["class"]=="LGBModel":
                         return True
                     return False

diff --git a/setup.py b/setup.py
@@ -56,6 +56,7 @@
     "joblib>=0.17.0",
     "ruamel.yaml>=0.16.12",
     "pymongo==3.7.2",  # For task management
+    "scikit-learn>=0.22",
 ]
 
 # Numpy include