add CI fix tool to app (#10)

* add CI fix tool to app

* perform ruff safe fix

* wrap too long lines in prompts.py

rdagent/factor_implementation/evolving/evaluators.py

@@ -8,6 +8,7 @@ from pandas.core.api import DataFrame as DataFrame
 from core.evolving_framework import Evaluator as EvolvingEvaluator
 from core.evolving_framework import Feedback, QueriedKnowledge
 from core.log import FinCoLog
+from core.utils import multiprocessing_wrapper
 from factor_implementation.evolving.evolvable_subjects import (
     FactorImplementationList,
 )
@@ -24,7 +25,6 @@ from factor_implementation.share_modules.factor import (
     FactorImplementation,
     FactorImplementationTask,
 )
-from core.utils import multiprocessing_wrapper
 
 
 class FactorImplementationSingleFeedback:

rdagent/factor_implementation/evolving/evolvable_subjects.py

@@ -1,7 +1,5 @@
 from __future__ import annotations
 
-import pandas as pd
-
 from core.evolving_framework import EvolvableSubjects
 from core.log import FinCoLog
 from factor_implementation.share_modules.factor import (

rdagent/factor_implementation/evolving/evolving_strategy.py

@@ -9,7 +9,7 @@ from typing import TYPE_CHECKING
 from jinja2 import Template
 
 from core.evolving_framework import EvolvingStrategy, QueriedKnowledge
-from oai.llm_utils import APIBackend
+from core.utils import multiprocessing_wrapper
 from factor_implementation.share_modules.conf import FactorImplementSettings
 from factor_implementation.share_modules.factor import (
     FactorImplementation,
@@ -20,7 +20,7 @@ from factor_implementation.share_modules.prompt import (
     FactorImplementationPrompts,
 )
 from factor_implementation.share_modules.utils import get_data_folder_intro
-from core.utils import multiprocessing_wrapper
+from oai.llm_utils import APIBackend
 
 if TYPE_CHECKING:
     from factor_implementation.evolving.evolvable_subjects import (

rdagent/factor_implementation/evolving/factor_implementation_evolving_cli.py

@@ -8,6 +8,7 @@ from fire.core import Fire
 from tqdm import tqdm
 
 from core.evolving_framework import EvoAgent, KnowledgeBase
+from core.utils import multiprocessing_wrapper
 from factor_implementation.evolving.evaluators import (
     FactorImplementationEvaluatorV1,
     FactorImplementationsMultiEvaluator,
@@ -29,7 +30,6 @@ from factor_implementation.share_modules.factor import (
     FactorImplementationTask,
     FileBasedFactorImplementation,
 )
-from core.utils import multiprocessing_wrapper
 
 ALPHA101_INIT_COMPONENTS = [
     "1. abs(): absolute value to certain columns",
@@ -108,10 +108,8 @@ class FactorImplementationEvolvingCli:
         if former_knowledge_base_path is not None and former_knowledge_base_path.exists():
             factor_knowledge_base = pickle.load(open(former_knowledge_base_path, "rb"))
             if self.evolving_version == 1 and not isinstance(
-                factor_knowledge_base, FactorImplementationKnowledgeBaseV1
-            ):
-                raise ValueError("The former knowledge base is not compatible with the current version")
-            elif self.evolving_version == 2 and not isinstance(
+                factor_knowledge_base, FactorImplementationKnowledgeBaseV1,
+            ) or self.evolving_version == 2 and not isinstance(
                 factor_knowledge_base,
                 FactorImplementationGraphKnowledgeBase,
             ):
@@ -261,7 +259,7 @@ class FactorImplementationEvolvingCli:
         print([feedback.final_decision if feedback is not None else None for feedback in feedbacks].count(True))
 
     def implement_amc(
-        self, evo_sub_path_str, former_knowledge_base_path_str, implementation_dump_path_str, slice_index
+        self, evo_sub_path_str, former_knowledge_base_path_str, implementation_dump_path_str, slice_index,
     ):
         factor_implementations: FactorImplementationList = pickle.load(open(evo_sub_path_str, "rb"))
         factor_implementations.target_factor_tasks = factor_implementations.target_factor_tasks[

rdagent/factor_implementation/evolving/knowledge_management.py

@@ -8,6 +8,7 @@ from itertools import combinations
 from pathlib import Path
 from typing import Union
 
+from finco.graph import UndirectedGraph, UndirectedNode
 from jinja2 import Template
 
 from core.evolving_framework import (
@@ -18,8 +19,6 @@ from core.evolving_framework import (
     QueriedKnowledge,
     RAGStrategy,
 )
-from finco.graph import UndirectedGraph, UndirectedNode
-from oai.llm_utils import APIBackend, calculate_embedding_distance_between_str_list
 from core.log import FinCoLog
 from factor_implementation.evolving.evaluators import (
     FactorImplementationSingleFeedback,
@@ -32,6 +31,7 @@ from factor_implementation.share_modules.factor import (
 from factor_implementation.share_modules.prompt import (
     FactorImplementationPrompts,
 )
+from oai.llm_utils import APIBackend, calculate_embedding_distance_between_str_list
 
 
 class FactorImplementationKnowledge(Knowledge):
@@ -150,47 +150,46 @@ class FactorImplementationRAGStrategyV1(RAGStrategy):
                 queried_knowledge.success_task_to_knowledge_dict[target_factor_task_information] = (
                     self.knowledgebase.implementation_trace[target_factor_task_information][-1]
                 )
+            elif (
+                len(
+                    self.knowledgebase.implementation_trace.setdefault(
+                        target_factor_task_information,
+                        [],
+                    ),
+                )
+                >= fail_task_trial_limit
+            ):
+                queried_knowledge.failed_task_info_set.add(target_factor_task_information)
             else:
-                if (
-                    len(
-                        self.knowledgebase.implementation_trace.setdefault(
-                            target_factor_task_information,
-                            [],
-                        ),
-                    )
-                    >= fail_task_trial_limit
-                ):
-                    queried_knowledge.failed_task_info_set.add(target_factor_task_information)
-                else:
-                    queried_knowledge.working_task_to_former_failed_knowledge_dict[target_factor_task_information] = (
-                        self.knowledgebase.implementation_trace.setdefault(
-                            target_factor_task_information,
-                            [],
-                        )[-v1_query_former_trace_limit:]
-                    )
+                queried_knowledge.working_task_to_former_failed_knowledge_dict[target_factor_task_information] = (
+                    self.knowledgebase.implementation_trace.setdefault(
+                        target_factor_task_information,
+                        [],
+                    )[-v1_query_former_trace_limit:]
+                )
 
-                    knowledge_base_success_task_list = list(
-                        self.knowledgebase.success_task_info_set,
-                    )
-                    similarity = calculate_embedding_distance_between_str_list(
-                        [target_factor_task_information],
-                        knowledge_base_success_task_list,
-                    )[0]
-                    similar_indexes = sorted(
-                        range(len(similarity)),
-                        key=lambda i: similarity[i],
-                        reverse=True,
-                    )[:v1_query_similar_success_limit]
-                    similar_successful_knowledge = [
-                        self.knowledgebase.implementation_trace.setdefault(
-                            knowledge_base_success_task_list[index],
-                            [],
-                        )[-1]
-                        for index in similar_indexes
-                    ]
-                    queried_knowledge.working_task_to_similar_successful_knowledge_dict[
-                        target_factor_task_information
-                    ] = similar_successful_knowledge
+                knowledge_base_success_task_list = list(
+                    self.knowledgebase.success_task_info_set,
+                )
+                similarity = calculate_embedding_distance_between_str_list(
+                    [target_factor_task_information],
+                    knowledge_base_success_task_list,
+                )[0]
+                similar_indexes = sorted(
+                    range(len(similarity)),
+                    key=lambda i: similarity[i],
+                    reverse=True,
+                )[:v1_query_similar_success_limit]
+                similar_successful_knowledge = [
+                    self.knowledgebase.implementation_trace.setdefault(
+                        knowledge_base_success_task_list[index],
+                        [],
+                    )[-1]
+                    for index in similar_indexes
+                ]
+                queried_knowledge.working_task_to_similar_successful_knowledge_dict[
+                    target_factor_task_information
+                ] = similar_successful_knowledge
         return queried_knowledge
 
 
@@ -889,7 +888,7 @@ class FactorImplementationGraphKnowledgeBase(KnowledgeBase):
             for possible_combination in possible_combinations:
                 node_list = list(possible_combination)
                 intersection_node_list.extend(
-                    self.graph.get_nodes_intersection(node_list, steps=steps, constraint_labels=constraint_labels)
+                    self.graph.get_nodes_intersection(node_list, steps=steps, constraint_labels=constraint_labels),
                 )
                 if output_intersection_origin:
                     for _ in range(len(intersection_node_list)):

rdagent/factor_implementation/share_modules/evaluator.py

@@ -3,10 +3,9 @@ from abc import ABC, abstractmethod
 from typing import Tuple
 
 import pandas as pd
+from finco.log import FinCoLog
 from jinja2 import Template
 
-from oai.llm_utils import APIBackend
-from finco.log import FinCoLog
 from factor_implementation.share_modules.conf import FactorImplementSettings
 from factor_implementation.share_modules.factor import (
     FactorImplementation,
@@ -15,6 +14,7 @@ from factor_implementation.share_modules.factor import (
 from factor_implementation.share_modules.prompt import (
     FactorImplementationPrompts,
 )
+from oai.llm_utils import APIBackend
 
 
 class Evaluator(ABC):
@@ -454,7 +454,7 @@ class FactorImplementationValueEvaluator(Evaluator):
                         )
 
                 except Exception as e:
-                    FinCoLog().warning(f"Error occurred when calculating the correlation: {str(e)}")
+                    FinCoLog().warning(f"Error occurred when calculating the correlation: {e!s}")
                     conclusions.append(
                         f"Some error occurred when calculating the correlation. Investigate the factors that might be causing the discrepancies and ensure that the logic of the factor calculation is consistent. Error: {e}",
                     )

rdagent/factor_implementation/share_modules/factor.py

@@ -7,15 +7,15 @@ from typing import Tuple, Union
 
 import pandas as pd
 from filelock import FileLock
-
-from oai.llm_utils import md5_hash
 from finco.log import FinCoLog
+
 from factor_implementation.share_modules.conf import FactorImplementSettings
 from factor_implementation.share_modules.exception import (
     CodeFormatException,
     NoOutputException,
     RuntimeErrorException,
 )
+from oai.llm_utils import md5_hash
 
 
 class FactorImplementationTask:
@@ -122,7 +122,7 @@ class FileBasedFactorImplementation(FactorImplementation):
                 raise ValueError(self.FB_CODE_NOT_SET)
         with FileLock(self.workspace_path / "execution.lock"):
             (Path.cwd() / "git_ignore_folder" / "factor_implementation_execution_cache").mkdir(
-                exist_ok=True, parents=True
+                exist_ok=True, parents=True,
             )
             if FactorImplementSettings().enable_execution_cache:
                 # NOTE: cache the result for the same code

rdagent/factor_implementation/share_modules/prompt.py

@@ -2,7 +2,6 @@ from pathlib import Path
 from typing import Dict
 
 import yaml
-
 from finco.utils import SingletonBaseClass