feat: add RD-Agent-Quant scenario (#838)

* fix model input shape bug and costeer_model bug * fix a bug * fix a bug in docker result extraction * a system-level optimization * add a filter of stdout * update * add stdout to model * model training_hyperparameters update * quant scenario * update some quant settings * llm choose action * Thompson Sampling Bandit for action choosing * refine both scens * add trace messages for quant scen * fix some bugs * fix some bugs * update * update * update * fix * fix * fix * update for merge * fix ci * fix some bugs * fix ci * fix ci * fix ci * fix ci * refactor * default qlib4rdagent local env downloading * fix ci * fix ci * fix a bug * fix ci * fix: align all prompts on template (#908) * use template to render all prompts * fix CI --------- Co-authored-by: Xu Yang <xuyang1@microsoft.com> * add fin_quant in cli * fix a bug * fix ci * fix some bugs * refactor * remove the columns in hypothesis if no value generated in this column * fix a bug * fix ci * fix conda env * add qlib gitignore * remove existed qlib folder & install torch in qlib conda * fix workspace ui in feedback * align model config in coder and runner in docker or conda * fix CI * fix CI --------- Co-authored-by: Xu Yang <peteryang@vip.qq.com> Co-authored-by: Xu Yang <xuyang1@microsoft.com>
2026-08-03 18:27:43 +00:00 · 2025-05-29 16:16:51 +08:00
parent b0e88c7375
commit d1019cb568
64 changed files with 2427 additions and 1025 deletions
@@ -1,9 +1,6 @@
 import json
-from pathlib import Path
 from typing import Dict

-from jinja2 import Environment, StrictUndefined
-
 from rdagent.components.coder.CoSTEER.config import CoSTEER_SETTINGS
 from rdagent.components.coder.CoSTEER.evaluators import CoSTEERSingleFeedback
 from rdagent.components.coder.CoSTEER.evolving_strategy import (
@@ -14,16 +11,13 @@ from rdagent.components.coder.CoSTEER.knowledge_management import (
    CoSTEERQueriedKnowledgeV2,
 )
 from rdagent.components.coder.model_coder.model import (
-    ModelExperiment,
    ModelFBWorkspace,
    ModelTask,
 )
 from rdagent.core.experiment import FBWorkspace
-from rdagent.core.prompts import Prompts
 from rdagent.oai.llm_conf import LLM_SETTINGS
 from rdagent.oai.llm_utils import APIBackend
-
-coder_prompts = Prompts(file_path=Path(__file__).parent / "prompts.yaml")
+from rdagent.utils.agent.tpl import T


 class ModelMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
@@ -52,31 +46,18 @@ class ModelMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
            if isinstance(queried_knowledge, CoSTEERQueriedKnowledgeV2)
            else queried_former_failed_knowledge
        )
-        system_prompt = (
-            Environment(undefined=StrictUndefined)
-            .from_string(
-                coder_prompts["evolving_strategy_model_coder"]["system"],
-            )
-            .render(
-                scenario=self.scen.get_scenario_all_desc(filtered_tag=target_task.model_type),
-                queried_former_failed_knowledge=queried_former_failed_knowledge_to_render,
-                current_code=workspace.file_dict.get("model.py"),
-            )
+        system_prompt = T(".prompts:evolving_strategy_model_coder.system").r(
+            scenario=self.scen.get_scenario_all_desc(filtered_tag="model"),
+            queried_former_failed_knowledge=queried_former_failed_knowledge_to_render,
+            current_code=workspace.file_dict.get("model.py"),
        )

        queried_similar_successful_knowledge_to_render = queried_similar_successful_knowledge
        for _ in range(10):  # max attempt to reduce the length of user_prompt
-            user_prompt = (
-                Environment(undefined=StrictUndefined)
-                .from_string(
-                    coder_prompts["evolving_strategy_model_coder"]["user"],
-                )
-                .render(
-                    model_information_str=model_information_str,
-                    queried_similar_successful_knowledge=queried_similar_successful_knowledge_to_render,
-                    queried_former_failed_knowledge=queried_former_failed_knowledge_to_render,
-                )
-                .strip("\n")
+            user_prompt = T(".prompts:evolving_strategy_model_coder.user").r(
+                model_information_str=model_information_str,
+                queried_similar_successful_knowledge=queried_similar_successful_knowledge_to_render,
+                queried_former_failed_knowledge=queried_former_failed_knowledge_to_render,
            )
            if (
                APIBackend().build_messages_and_calculate_token(