use mp to execute all feature and fix a small bug in config (#157)

rdagent/app/qlib_rd_loop/factor_from_report_sh.py

@@ -114,6 +114,7 @@ class FactorReportLoop(LoopBase, metaclass=LoopMeta):
                 if self.pdf_file_index > 100:
                     break
                 report_file_path = self.judge_pdf_data_items[self.pdf_file_index]
+                logger.info(f"Processing number {self.pdf_file_index} report: {report_file_path}")
                 self.pdf_file_index += 1
                 exp, hypothesis = extract_hypothesis_and_exp_from_reports(str(report_file_path))
                 if exp is None:
@@ -153,7 +154,7 @@ def main(path=None, step_n=None):
 
     .. code-block:: python
 
-        dotenv run -- python rdagent/app/qlib_rd_loop/factor_from_report_sh.py $LOG_PATH/__session__/1/0_propose  --step_n 1   # `step_n` is a optional paramter
+        dotenv run -- python rdagent/app/qlib_rd_loop/factor_from_report_sh.py $LOG_PATH/__session__/1/0_propose  --step_n 1   # `step_n` is a optional parameter
 
     """
     if path is None:

rdagent/components/coder/factor_coder/CoSTEER/__init__.py

@@ -24,6 +24,7 @@ from rdagent.components.coder.factor_coder.factor import FactorExperiment
 from rdagent.core.developer import Developer
 from rdagent.core.evolving_agent import RAGEvoAgent
 from rdagent.core.scenario import Scenario
+from rdagent.log import rdagent_logger as logger
 
 
 class FactorCoSTEER(Developer[FactorExperiment]):
@@ -107,5 +108,6 @@ class FactorCoSTEER(Developer[FactorExperiment]):
         # save new knowledge base
         if self.new_knowledge_base_path is not None:
             pickle.dump(factor_knowledge_base, open(self.new_knowledge_base_path, "wb"))
+            logger.info(f"New knowledge base saved to {self.new_knowledge_base_path}")
         exp.sub_workspace_list = factor_experiment.sub_workspace_list
         return exp

rdagent/components/coder/factor_coder/config.py

@@ -44,7 +44,7 @@ class FactorImplementSettings(BaseSettings):
     max_loop: int = 10
 
     knowledge_base_path: Union[str, None] = None
-    new_knowledge_base_path: Union[str, None] = knowledge_base_path
+    new_knowledge_base_path: Union[str, None] = None
 
     python_bin: str = "python"
 

rdagent/components/coder/model_coder/conf.py

@@ -15,7 +15,7 @@ class ModelImplSettings(BaseSettings):
     )
 
     knowledge_base_path: Union[str, None] = None
-    new_knowledge_base_path: Union[str, None] = knowledge_base_path
+    new_knowledge_base_path: Union[str, None] = None
 
     max_loop: int = 10
 

rdagent/scenarios/qlib/developer/factor_runner.py

@@ -5,6 +5,9 @@ from typing import List
 import pandas as pd
 from pandarallel import pandarallel
 
+from rdagent.core.conf import RD_AGENT_SETTINGS
+from rdagent.core.utils import multiprocessing_wrapper
+
 pandarallel.initialize(verbose=1)
 
 from rdagent.components.runner import CachedRunner
@@ -139,9 +142,11 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
         # Collect all exp's dataframes
         for exp in exp_or_list:
             # Iterate over sub-implementations and execute them to get each factor data
-            for implementation in exp.sub_workspace_list:
-                message, df = implementation.execute(data_type="All")
-
+            message_and_df_list = multiprocessing_wrapper(
+                [(implementation.execute, (False, "All")) for implementation in exp.sub_workspace_list],
+                n=RD_AGENT_SETTINGS.multi_proc_n,
+            )
+            for message, df in message_and_df_list:
                 # Check if factor generation was successful
                 if df is not None and "datetime" in df.index.names:
                     time_diff = df.index.get_level_values("datetime").to_series().diff().dropna().unique()