feat: make spec optional (#719)

* feat: Add spec_enabled configuration for data science settings

* make spec alternative

* change spec logic in exp_gen

* remove some general texts

* align

---------

Co-authored-by: Young <afe.young@gmail.com>
Co-authored-by: yuanteli <1957922024@qq.com>

rdagent/components/coder/data_science/model/__init__.py

@@ -1,5 +1,7 @@
+from pathlib import Path
 from typing import Dict
 
+from rdagent.app.data_science.conf import DS_RD_SETTING
 from rdagent.components.coder.CoSTEER import CoSTEER
 from rdagent.components.coder.CoSTEER.evaluators import (
     CoSTEERMultiEvaluator,
@@ -23,6 +25,8 @@ from rdagent.oai.llm_utils import APIBackend
 from rdagent.utils.agent.ret import PythonBatchEditOut
 from rdagent.utils.agent.tpl import T
 
+DIRNAME = Path(__file__).absolute().resolve().parent
+
 
 class ModelMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
     def implement_one_task(
@@ -71,8 +75,16 @@ class ModelMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
         #     latest_code=workspace.file_dict.get(f"{target_task.name}.py", None),
         # )
         # We want to use a simpler way to
+        code_spec = (
+            workspace.file_dict["spec/model.md"]
+            if DS_RD_SETTING.spec_enabled
+            else T("scenarios.data_science.share:component_spec.general").r(
+                spec=T("scenarios.data_science.share:component_spec.Model").r(),
+                test_code=(DIRNAME / "eval_tests" / "model_test.txt").read_text().replace("model01", target_task.name),
+            )
+        )
         user_prompt = T(".prompts:model_coder.user_general").r(
-            model_spec=workspace.file_dict["spec/model.md"],
+            code_spec=code_spec,
             latest_model_code=workspace.get_codes(
                 r"^model_(?!test)\w+\.py$"
             ),  # TODO: If we have high failure rate here, we should clean this step with less information.

rdagent/components/coder/data_science/model/eval.py

@@ -98,8 +98,6 @@ class ModelGeneralCaseSpecEvaluator(CoSTEEREvaluator):
                 task_desc=target_task.get_task_information(),
                 test_code=test_code,
                 code=implementation.file_dict[f"{target_task.name}.py"],
-                scenario=self.scen.get_scenario_all_desc(),
-                spec=implementation.file_dict["spec/model.md"],
                 workflow_stdout=workflow_stdout,
                 workflow_code=implementation.all_codes,
             )

rdagent/components/coder/data_science/model/eval_tests/model_test.txt

@@ -1,3 +1,6 @@
+"""
+Tests for `model_workflow` in model01.py
+"""
 import time
 
 from feature import feat_eng

rdagent/components/coder/data_science/model/prompts.yaml

@@ -53,8 +53,8 @@ model_coder:
     {% endif %}
 
   user_general: |-
-    --------- Model Building Specification ---------
-    {{ model_spec }}
+    --------- Code Specification ---------
+    {{ code_spec }}
 
     --------- Former model code ---------
     {% if latest_model_code|length == 0 %}