From fb8f2b561eeeba7fc375ef9fec63f2a7c636d3d5 Mon Sep 17 00:00:00 2001 From: Yuante Li <104308117+WinstonLiyt@users.noreply.github.com> Date: Wed, 2 Jul 2025 23:31:29 +0800 Subject: [PATCH] fix: fix a minor bug in DS eval (#1012) --- rdagent/components/coder/data_science/ensemble/eval.py | 2 +- rdagent/components/coder/data_science/feature/eval.py | 4 ++-- rdagent/components/coder/data_science/model/eval.py | 4 ++-- rdagent/components/coder/data_science/pipeline/eval.py | 4 ++-- rdagent/components/coder/data_science/raw_data_loader/eval.py | 2 +- rdagent/components/coder/data_science/workflow/eval.py | 2 +- rdagent/scenarios/data_science/dev/runner/eval.py | 2 +- rdagent/scenarios/data_science/test_eval.py | 4 ++-- 8 files changed, 12 insertions(+), 12 deletions(-) diff --git a/rdagent/components/coder/data_science/ensemble/eval.py b/rdagent/components/coder/data_science/ensemble/eval.py index d844e17d..fd51f159 100644 --- a/rdagent/components/coder/data_science/ensemble/eval.py +++ b/rdagent/components/coder/data_science/ensemble/eval.py @@ -65,7 +65,7 @@ class EnsembleCoSTEEREvaluator(CoSTEEREvaluator): implementation.inject_files(**{fname: test_code}) result = implementation.run(env=env, entry=f"python {fname}") stdout = result.stdout - ret_code = result.ret_code + ret_code = result.exit_code stdout += f"\nNOTE: the above scripts run with return code {ret_code}" diff --git a/rdagent/components/coder/data_science/feature/eval.py b/rdagent/components/coder/data_science/feature/eval.py index 18b79a49..e076fcf2 100644 --- a/rdagent/components/coder/data_science/feature/eval.py +++ b/rdagent/components/coder/data_science/feature/eval.py @@ -52,7 +52,7 @@ class FeatureCoSTEEREvaluator(CoSTEEREvaluator): result = implementation.run(env=env, entry=f"python {fname}") - if "main.py" in implementation.file_dict and result.ret_code == 0: + if "main.py" in implementation.file_dict and result.exit_code == 0: workflow_stdout = implementation.execute(env=env, entry="python main.py") workflow_stdout = remove_eda_part(workflow_stdout) else: @@ -76,6 +76,6 @@ class FeatureCoSTEEREvaluator(CoSTEEREvaluator): user_prompt=user_prompt, init_kwargs_update_func=FeatureEvalFeedback.val_and_update_init_dict, ) - fb.final_decision = fb.final_decision and result.ret_code == 0 + fb.final_decision = fb.final_decision and result.exit_code == 0 return fb diff --git a/rdagent/components/coder/data_science/model/eval.py b/rdagent/components/coder/data_science/model/eval.py index 75e21133..ea429af4 100644 --- a/rdagent/components/coder/data_science/model/eval.py +++ b/rdagent/components/coder/data_science/model/eval.py @@ -69,7 +69,7 @@ class ModelGeneralCaseSpecEvaluator(CoSTEEREvaluator): implementation.inject_files(**{fname: test_code}) result = implementation.run(env=env, entry=f"python {fname}") stdout = result.stdout - ret_code = result.ret_code + ret_code = result.exit_code if stdout is None: raise CoderError( @@ -115,6 +115,6 @@ class ModelGeneralCaseSpecEvaluator(CoSTEEREvaluator): user_prompt=user_prompt, init_kwargs_update_func=ModelSingleFeedback.val_and_update_init_dict, ) - fb.final_decision = fb.final_decision and result.ret_code == 0 + fb.final_decision = fb.final_decision and result.exit_code == 0 return fb diff --git a/rdagent/components/coder/data_science/pipeline/eval.py b/rdagent/components/coder/data_science/pipeline/eval.py index e7e3e8df..19b25bf4 100644 --- a/rdagent/components/coder/data_science/pipeline/eval.py +++ b/rdagent/components/coder/data_science/pipeline/eval.py @@ -59,7 +59,7 @@ class PipelineCoSTEEREvaluator(CoSTEEREvaluator): implementation.execute(env=env, entry=get_clear_ws_cmd()) result = implementation.run(env=env, entry=f"python -m coverage run main.py") implementation.running_info.running_time = result.running_time - execute_ret_code = result.ret_code + execute_ret_code = result.exit_code stdout = remove_eda_part(result.stdout) stdout += f"The code executed {'successfully' if execute_ret_code == 0 else 'failed'}." @@ -109,7 +109,7 @@ class PipelineCoSTEEREvaluator(CoSTEEREvaluator): # stdout += "----Submission Check 1-----\n" submission_result = implementation.run(env=env, entry="python test/submission_format_test.py") submission_check_out = submission_result.stdout - submission_ret_code = submission_result.ret_code + submission_ret_code = submission_result.exit_code if DS_RD_SETTING.rule_base_eval: if execute_ret_code == 0 and score_ret_code == 0 and submission_ret_code == 0: return PipelineSingleFeedback( diff --git a/rdagent/components/coder/data_science/raw_data_loader/eval.py b/rdagent/components/coder/data_science/raw_data_loader/eval.py index 69154e3a..037979e2 100644 --- a/rdagent/components/coder/data_science/raw_data_loader/eval.py +++ b/rdagent/components/coder/data_science/raw_data_loader/eval.py @@ -54,7 +54,7 @@ class DataLoaderCoSTEEREvaluator(CoSTEEREvaluator): implementation.inject_files(**{fname: test_code}) result = implementation.run(env=env, entry=f"python {fname}") stdout = result.stdout - ret_code = result.ret_code + ret_code = result.exit_code match = re.search(r"(.*?)=== Start of EDA part ===(.*)=== End of EDA part ===(.*)", stdout, re.DOTALL) stdout_part_1, eda_output, stdout_part_2 = match.groups() if match else (stdout, None, "") stdout = stdout_part_1 + stdout_part_2 diff --git a/rdagent/components/coder/data_science/workflow/eval.py b/rdagent/components/coder/data_science/workflow/eval.py index 72798c40..1731ce3e 100644 --- a/rdagent/components/coder/data_science/workflow/eval.py +++ b/rdagent/components/coder/data_science/workflow/eval.py @@ -123,7 +123,7 @@ class WorkflowGeneralCaseSpecEvaluator(CoSTEEREvaluator): # stdout += "----Submission Check 1-----\n" submission_result = implementation.run(env=env, entry="python test/submission_format_test.py") submission_check_out = submission_result.stdout - submission_ret_code = submission_result.ret_code + submission_ret_code = submission_result.exit_code stdout += "\n" + submission_check_out system_prompt = T(".prompts:workflow_eval.system").r( diff --git a/rdagent/scenarios/data_science/dev/runner/eval.py b/rdagent/scenarios/data_science/dev/runner/eval.py index 3f85269a..9ce9ca9e 100644 --- a/rdagent/scenarios/data_science/dev/runner/eval.py +++ b/rdagent/scenarios/data_science/dev/runner/eval.py @@ -55,7 +55,7 @@ class DSCoSTEERCoSTEEREvaluator(CoSTEEREvaluator): # execute workflow result = implementation.run(env=env, entry="python -m coverage run main.py") stdout = result.stdout - execute_ret_code = result.ret_code + execute_ret_code = result.exit_code implementation.running_info.running_time = result.running_time match = re.search(r"(.*?)=== Start of EDA part ===(.*)=== End of EDA part ===", stdout, re.DOTALL) diff --git a/rdagent/scenarios/data_science/test_eval.py b/rdagent/scenarios/data_science/test_eval.py index 9db0b8ec..8c2141cf 100644 --- a/rdagent/scenarios/data_science/test_eval.py +++ b/rdagent/scenarios/data_science/test_eval.py @@ -82,7 +82,7 @@ class TestEval(TestEvalBase): **{file: workspace.DEL_KEY for file in ["submission_format_valid.py", "submission_test.csv"]} ) workspace.inject_files(**{"test/mle_submission_format_test.output": submission_result.stdout}) - return submission_result.stdout, submission_result.ret_code + return submission_result.stdout, submission_result.exit_code def enabled(self, competition) -> bool: return Path( @@ -119,7 +119,7 @@ class MLETestEval(TestEvalBase): submission_result = workspace.run(env=self.env, entry="python test/mle_submission_format_test.py") workspace.inject_files(**{"test/mle_submission_format_test.output": submission_result.stdout}) - return submission_result.stdout, submission_result.ret_code + return submission_result.stdout, submission_result.exit_code def enabled(self, competition) -> bool: return True