Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

chore: change log object #553

Open
wants to merge 6 commits into
base: main
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
182 changes: 182 additions & 0 deletions rdagent/log/ui/utils.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,182 @@
from collections import defaultdict
from datetime import datetime, timezone
from pathlib import Path

import plotly

from rdagent.components.coder.CoSTEER.evaluators import CoSTEERSingleFeedbackDeprecated
from rdagent.components.coder.factor_coder.evaluators import FactorSingleFeedback
from rdagent.components.coder.factor_coder.factor import FactorFBWorkspace, FactorTask
from rdagent.components.coder.model_coder.model import ModelFBWorkspace, ModelTask
from rdagent.core.proposal import Hypothesis, HypothesisFeedback
from rdagent.log.storage import FileStorage
from rdagent.log.ui.llm_st import extract_evoid, extract_loopid_func_name
from rdagent.log.ui.qlib_report_figure import report_figure

msgs_for_frontend = defaultdict(list)


def format_pkl(base_path: str | Path):
base_path = Path(base_path)
dir2id = {dir_name.name: idx for idx, dir_name in enumerate(base_path.iterdir())}
for dn, did in dir2id.items():
fs = FileStorage(base_path / dn)
for m in fs.iter_msg():
lp = extract_loopid_func_name(m.tag)
lp_id = lp[0] if lp and lp[0] is not None else None
# lp_id = (lp := extract_loopid_func_name(m.tag))[0] if lp[0] is not None else None
if "r.hypothesis generation" in m.tag:
h: Hypothesis = m.content
msgs_for_frontend[did].append(
{
"tag": "research.hypothesis",
"timestamp": m.timestamp.isoformat(),
"content": {
"name_map": {
"hypothesis": "RD-Agent proposes the hypothesis⬇️",
"concise_justification": "because the reason⬇️",
"concise_observation": "based on the observation⬇️",
"concise_knowledge": "Knowledge⬇️ gained after practice",
},
"hypothesis": h.hypothesis,
"concise_justification": h.concise_justification,
"concise_observation": h.concise_observation,
"concise_knowledge": h.concise_knowledge,
},
}
)

# m.tag 中不存在 d.load_experiment, 存在 r.load_experiment
elif "r.experiment generation" in m.tag or "d.load_experiment" in m.tag:
if "d.load_experiment" in m.tag:
tasks: list[FactorTask | ModelTask] = m.content.sub_tasks
else:
tasks: list[FactorTask | ModelTask] = m.content
if isinstance(tasks[0], FactorTask):
msgs_for_frontend[did].append(
{
"tag": "research.tasks",
"timestamp": m.timestamp.isoformat(),
"content": [
{
"name": t.factor_name,
"description": t.factor_description,
"formulation": t.factor_formulation,
"variables": t.variables,
}
for t in tasks
],
}
)
elif isinstance(tasks[0], ModelTask):
msgs_for_frontend[did].append(
{
"tag": "research.tasks",
"timestamp": m.timestamp.isoformat(),
"content": [
{
"name": t.name,
"description": t.description,
"model_type": t.model_type,
"formulation": t.formulation,
"variables": t.variables,
}
for t in tasks
],
}
)

elif f"evo_loop_{lp_id}.evolving code" in m.tag:
ws: list[FactorFBWorkspace | ModelFBWorkspace] = [i for i in m.content]
msgs_for_frontend[did].append(
{
"tag": "evolving.codes",
"timestamp": m.timestamp.isoformat(),
"content": [
{
"target_task_name": (
w.target_task.name
if isinstance(w.target_task, ModelTask)
else w.target_task.factor_name
),
"code": w.file_dict,
}
for w in ws
if w
],
}
)

elif f"evo_loop_{lp_id}.evolving feedback" in m.tag:
fl: list[FactorSingleFeedback | CoSTEERSingleFeedbackDeprecated] = [i for i in m.content]
msgs_for_frontend[did].append(
{
"tag": "evolving.feedbacks",
"timestamp": m.timestamp.isoformat(),
"content": [
{
"final_decision": f.final_decision,
"final_feedback": f.final_feedback,
"execution_feedback": f.execution_feedback,
"code_feedback": f.code_feedback,
"value_feedback": (
f.value_feedback
if isinstance(f, CoSTEERSingleFeedbackDeprecated)
else f.factor_value_feedback
),
"model_shape_feedback": (
f.shape_feedback if isinstance(f, CoSTEERSingleFeedbackDeprecated) else None
),
}
for f in fl
if f
],
}
)

elif "scenario" in m.tag:
msgs_for_frontend[did].append(
{
"tag": "feedback.config",
"timestamp": m.timestamp.isoformat(),
"content": {"config": m.content.experiment_setting},
}
)

elif "ef.Quantitative Backtesting Chart" in m.tag:
msgs_for_frontend[did].append(
{
"tag": "feedback.return_chart",
"timestamp": m.timestamp.isoformat(),
"content": {"chart_html": plotly.io.to_html(report_figure(m.content))},
}
)

elif "model runner result" in m.tag or "factor runner result" in m.tag or "runner result" in m.tag:
msgs_for_frontend[did].append(
{
"tag": "feedback.metric",
"timestamp": m.timestamp.isoformat(),
"content": {"result": m.content.result.iloc[0]},
}
)

elif "ef.feedback" in m.tag:
hf: HypothesisFeedback = m.content
msgs_for_frontend[did].append(
{
"tag": "feedback.hypothesis_feedback",
"timestamp": m.timestamp.isoformat(),
"content": {
"observations": hf.observations,
"hypothesis_evaluation": hf.hypothesis_evaluation,
"new_hypothesis": hf.new_hypothesis,
"decision": hf.decision,
"reason": hf.reason,
},
}
)
for msgs in msgs_for_frontend.values():
msgs.append({"tag": "END", "timestamp": datetime.now(timezone.utc).isoformat(), "content": {}})

return msgs_for_frontend
1 change: 1 addition & 0 deletions test/utils/test_import.py
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,7 @@ def import_all_modules_from_directory(directory):
or fstr.endswith("rdagent/app/cli.py")
or fstr.endswith("rdagent/app/CI/run.py")
or fstr.endswith("rdagent/app/utils/ape.py")
or fstr.endswith("rdagent/log/ui/utils.py")
):
# the entrance points
continue
Expand Down