gavindiaz/NexQuant
1
0
Fork 0
mirror of https://github.com/NicolasBohn/NexQuant.git synced 2026-07-27 23:47:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: add DSExperiment type check and directory validation in log proc… (#535)

Browse Source 
* fix: Add DSExperiment type check and directory validation in log processing

* style: Reformat FileNotFoundError message for readability
This commit is contained in:
you-n-g
2025-01-24 00:03:41 +08:00
committed by GitHub
parent a290f3cee9
commit 2857200d2a
2 changed files with 29 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files
rdagent/log/mle_summary.py
+28 -23
View File
@@ -8,6 +8,7 @@ import pandas as pd
from rdagent.app.data_science.conf import DS_RD_SETTING
from rdagent.log.storage import FileStorage
from rdagent.scenarios.data_science.experiment.experiment import DSExperiment
from rdagent.utils.env import DockerEnv, MLEBDockerConf
mle_de_conf = MLEBDockerConf()
@@ -31,16 +32,18 @@ def save_grade_info(log_trace_path: Path):
competition = msg.content
if "running" in msg.tag:
msg.content.experiment_workspace.execute(
env=de,
entry=f"bash -c 'mlebench grade-sample submission.csv {competition} --data-dir /mle/data > mle_score.txt 2>&1'",
)
msg.content.experiment_workspace.execute(env=de, entry="chmod 777 mle_score.txt")
if isinstance(msg.content, DSExperiment):
msg.content.experiment_workspace.execute(
env=de,
entry=f"bash -c 'mlebench grade-sample submission.csv {competition} --data-dir /mle/data > mle_score.txt 2>&1'",
)
msg.content.experiment_workspace.execute(env=de, entry="chmod 777 mle_score.txt")
def save_all_grade_info(log_folder):
for log_trace_path in log_folder.iterdir():
save_grade_info(log_trace_path)
if log_trace_path.is_dir():
save_grade_info(log_trace_path)
def summarize_folder(log_folder: Path):
@@ -62,24 +65,26 @@ def summarize_folder(log_folder: Path):
if "direct_exp_gen" in msg.tag:
loop_num += 1
if "running" in msg.tag:
submission_path = msg.content.experiment_workspace.workspace_path / "submission.csv"
if submission_path.exists():
made_submission_num += 1
scores_path = msg.content.experiment_workspace.workspace_path / "scores.csv"
valid_scores[loop_num - 1] = pd.read_csv(scores_path, index_col=0)
grade_output_path = msg.content.experiment_workspace.workspace_path / "mle_score.txt"
if not grade_output_path.exists():
raise FileNotFoundError(f"mle_score.txt in {grade_output_path} not found, genarate it first!")
grade_output = extract_mle_json(grade_output_path.read_text())
if grade_output["score"] is not None:
test_scores[loop_num - 1] = grade_output["score"]
if grade_output["any_medal"]:
medal = (
"gold"
if grade_output["gold_medal"]
else "silver" if grade_output["silver_medal"] else "bronze"
if isinstance(msg.content, DSExperiment):
submission_path = msg.content.experiment_workspace.workspace_path / "submission.csv"
if submission_path.exists():
made_submission_num += 1
scores_path = msg.content.experiment_workspace.workspace_path / "scores.csv"
valid_scores[loop_num - 1] = pd.read_csv(scores_path, index_col=0)
grade_output_path = msg.content.experiment_workspace.workspace_path / "mle_score.txt"
if not grade_output_path.exists():
raise FileNotFoundError(
f"mle_score.txt in {grade_output_path} not found, genarate it first!"
)
grade_output = extract_mle_json(grade_output_path.read_text())
if grade_output["score"] is not None:
test_scores[loop_num - 1] = grade_output["score"]
if grade_output["any_medal"]:
medal = (
"gold"
if grade_output["gold_medal"]
else "silver" if grade_output["silver_medal"] else "bronze"
)
if "feedback" in msg.tag and "evolving" not in msg.tag:
if bool(msg.content):
rdagent/utils/workflow.py
+1 -1
View File
@@ -109,8 +109,8 @@ class LoopBase:
step_n -= 1
li, si = self.loop_idx, self.step_idx
name = self.steps[si]
logger.info(f"Start Loop {li}, Step {si}: {name}")
with logger.tag(f"Loop_{li}.{name}"):
start = datetime.datetime.now(datetime.timezone.utc)
func = getattr(self, name)