feat: Factor Implement Search Enhancement (#294)

* Search enhancement

* refactor: reorganize imports for consistency with isort

* reformatterd by black

---------

Co-authored-by: Tim <illking@foxmail.com>

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

@@ -1,3 +1,4 @@
+import json
 import pickle
 from pathlib import Path
 
@@ -49,6 +50,11 @@ class FactorCoSTEER(Developer[FactorExperiment]):
             if FACTOR_IMPLEMENT_SETTINGS.new_knowledge_base_path is not None
             else None
         )
+        self.data_tables_knowledge_path = (
+            Path(FACTOR_IMPLEMENT_SETTINGS.data_tables_knowledge_path)
+            if FACTOR_IMPLEMENT_SETTINGS.data_tables_knowledge_path is not None
+            else None
+        )
         self.with_knowledge = with_knowledge
         self.with_feedback = with_feedback
         self.knowledge_self_gen = knowledge_self_gen
@@ -72,6 +78,7 @@ class FactorCoSTEER(Developer[FactorExperiment]):
             factor_knowledge_base = (
                 FactorGraphKnowledgeBase(
                     init_component_list=component_init_list,
+                    data_set_knowledge_path=self.data_tables_knowledge_path,
                 )
                 if self.evolving_version == 2
                 else FactorKnowledgeBaseV1()

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

@@ -183,6 +183,19 @@ class FactorEvolvingStrategyWithGraph(MultiProcessEvolvingStrategy):
         self.num_loop = 0
         self.haveSelected = False
 
+    def _query_data_tables(self, user_prompt, session):
+        for _ in range(10):  # max attempt to reduce the length of user_prompt
+            response = session.build_chat_completion(
+                user_prompt=user_prompt,
+                json_mode=True,
+            )
+            try:
+                result = json.loads(response)
+                return result
+            except json.JSONDecodeError:
+                continue
+        return None
+
     def implement_one_factor(
         self,
         target_task: FactorTask,
@@ -218,6 +231,42 @@ class FactorEvolvingStrategyWithGraph(MultiProcessEvolvingStrategy):
                 queried_knowledge.former_traces[target_factor_task_information] if queried_knowledge is not None else []
             )
 
+            queried_data_tables = (
+                queried_knowledge.data_set_knowledge_dict[target_factor_task_information]
+                if queried_knowledge is not None
+                else []
+            )
+            queried_data_tables_str = json.dumps(queried_data_tables, indent=2)
+
+            system_prompt = (
+                Environment(undefined=StrictUndefined)
+                .from_string(
+                    implement_prompts["evolving_strategy_search_data_table_system_prompt"],
+                )
+                .render()
+            )
+
+            user_prompt = (
+                Environment(undefined=StrictUndefined)
+                .from_string(
+                    implement_prompts["evolving_strategy_search_data_table"],
+                )
+                .render(
+                    scenario=self.scen.get_scenario_all_desc(),
+                    factor_information_str=target_factor_task_information,
+                    data_tables=queried_data_tables_str,
+                )
+            )
+            session = APIBackend(use_chat_cache=FACTOR_IMPLEMENT_SETTINGS.coder_use_cache).build_chat_session(
+                session_system_prompt=system_prompt,
+            )
+
+            useful_data_table = self._query_data_tables(user_prompt, session)
+            selected_knowledge_dict = {}
+            for key in useful_data_table:
+                if key in queried_knowledge.data_set_knowledge_dict:
+                    selected_knowledge_dict[key] = queried_knowledge.data_set_knowledge_dict[key]
+
             queried_former_failed_knowledge_to_render = queried_former_failed_knowledge
 
             system_prompt = (
@@ -228,6 +277,7 @@ class FactorEvolvingStrategyWithGraph(MultiProcessEvolvingStrategy):
                 .render(
                     scenario=self.scen.get_scenario_all_desc(),
                     queried_former_failed_knowledge=queried_former_failed_knowledge_to_render,
+                    selected_knowledge_dict=selected_knowledge_dict,
                 )
             )
 

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

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import copy
+import heapq
 import json
 import random
 import re
@@ -204,11 +205,13 @@ class FactorQueriedGraphKnowledge(FactorQueriedKnowledge):
         former_traces: dict = {},
         component_with_success_task: dict = {},
         error_with_success_task: dict = {},
+        data_set_knowledge_dict: dict = {},
         **kwargs,
     ) -> None:
         self.former_traces = former_traces
         self.component_with_success_task = component_with_success_task
         self.error_with_success_task = error_with_success_task
+        self.data_set_knowledge_dict = data_set_knowledge_dict
         super().__init__(**kwargs)
 
 
@@ -308,6 +311,10 @@ class FactorGraphRAGStrategy(RAGStrategy):
             FACTOR_IMPLEMENT_SETTINGS.v2_query_error_limit,
             knowledge_sampler=conf_knowledge_sampler,
         )
+        factor_implementation_queried_graph_knowledge = self.dataset_query(
+            evo,
+            factor_implementation_queried_graph_knowledge,
+        )
         return factor_implementation_queried_graph_knowledge
 
     def analyze_component(
@@ -710,9 +717,37 @@ class FactorGraphRAGStrategy(RAGStrategy):
 
         return factor_implementation_queried_graph_knowledge
 
+    def dataset_query(
+        self,
+        evo: EvolvableSubjects,
+        factor_implementation_queried_graph_knowledge: FactorQueriedGraphKnowledge,
+    ) -> QueriedKnowledge | None:
+        for task_index, target_factor_task in enumerate(evo.sub_tasks):
+            target_factor_task_information = target_factor_task.get_task_information()
+            related_info = {}
+
+            knowledge_dict = self.knowledgebase.data_set_knowledge_dict
+            table_explanations = [f"{key}: {json.dumps(value)}" for key, value in knowledge_dict.items()]
+
+            similarity = calculate_embedding_distance_between_str_list(
+                [target_factor_task_information], table_explanations
+            )[0]
+
+            top_related_indexes = heapq.nlargest(10, range(len(similarity)), key=lambda i: similarity[i])
+
+            for index in top_related_indexes:
+                key = list(knowledge_dict.keys())[index]
+                related_info[key] = knowledge_dict[key]
+
+            factor_implementation_queried_graph_knowledge.data_set_knowledge_dict[
+                target_factor_task_information
+            ] = related_info
+
+        return factor_implementation_queried_graph_knowledge
+
 
 class FactorGraphKnowledgeBase(KnowledgeBase):
-    def __init__(self, init_component_list=None) -> None:
+    def __init__(self, init_component_list=None, data_set_knowledge_path=None) -> None:
         """
         Load knowledge, offer brief information of knowledge and common handle interfaces
         """
@@ -740,6 +775,12 @@ class FactorGraphKnowledgeBase(KnowledgeBase):
         # store the task description to component nodes
         self.task_to_component_nodes = {}
 
+        # data set: data set information
+        self.data_set_knowledge_dict = {}
+        if data_set_knowledge_path:
+            with open(data_set_knowledge_path, "r") as f:
+                self.data_set_knowledge_dict = json.load(f)
+
     def get_all_nodes_by_label(self, label: str) -> list[UndirectedNode]:
         return self.graph.get_all_nodes_by_label(label)
 

rdagent/components/coder/factor_coder/config.py

@@ -55,6 +55,9 @@ class FactorImplementSettings(BaseSettings):
     knowledge_base_path: Union[str, None] = None
     """Path to the knowledge base"""
 
+    data_tables_knowledge_path: Union[str, None] = None
+    """Path to the data tables knowledge"""
+
     new_knowledge_base_path: Union[str, None] = None
     """Path to the new knowledge base"""
 

rdagent/components/coder/factor_coder/prompts.yaml

@@ -46,6 +46,7 @@ evolving_strategy_factor_implementation_v1_system: |-
   1. The user might provide you the correct code to similar factors. Your should learn from these code to write the correct code.
   2. The user might provide you the failed former code and the corresponding feedback to the code. The feedback contains to the execution, the code and the factor value. You should analyze the feedback and try to correct the latest code.
   3. The user might provide you the suggestion to the latest fail code and some similar fail to correct pairs. Each pair contains the fail code with similar error and the corresponding corrected version code. You should learn from these suggestion to write the correct code.
+  4. The user might provide you some data table names and their explanations, as well as their corresponding data fields. You should find the data fields that contains the data needed for the factor implementation.
 
   Your must write your code based on your former latest attempt below which consists of your former code and code feedback, you should read the former attempt carefully and must not modify the right part of your former code.
 
@@ -57,6 +58,13 @@ evolving_strategy_factor_implementation_v1_system: |-
   {{ queried_former_failed_knowledge[-1].feedback }}
   {% endif %}
 
+  {% if selected_knowledge_dict|length != 0 %}
+  --------------Some datasets and their fields you may need:---------------
+  {% for selected_knowledge in selected_knowledge_dict.keys() %}
+  {{ selected_knowledge }}: {{ selected_knowledge_dict[selected_knowledge] }}
+  {% endfor %}
+  {% endif %}
+
   Please response the code in the following json format. Here is an example structure for the JSON output:
   {
       "code": "The Python code as a string."
@@ -119,6 +127,27 @@ evolving_strategy_factor_implementation_v2_user: |-
   {% endfor %}
   {% endif %}
 
+evolving_strategy_search_data_table_system_prompt: |-
+  The user will provide you with a task and some data. You need to determine which data tables are needed by the user.
+
+  Respond with your analysis in JSON format. The JSON schema should include:
+    {
+        "name of data table 1": ["name of data field 1", "name of data field 2"],
+        "name of data table 2": ["name of data field 1"],
+    }
+
+evolving_strategy_search_data_table: |-
+  User is trying to implement some factors in the following scenario:
+    {{ scenario }}
+    User is doing the following task: 
+    {{factor_information_str}}
+
+    Here are some data table names and their explanations. 
+    You should find the data table and data columns that contains the data needed for the factor implementation. 
+    Please response the data table name in the json format.
+    -------------------Data Tables-------------------
+    {{ data_tables }}
+
 evolving_strategy_error_summary_v2_system: |-
   User is trying to implement some factors in the following scenario:
   {{ scenario }}