gavindiaz/NexQuant
1
0
Fork 0
mirror of https://github.com/NicolasBohn/NexQuant.git synced 2026-08-09 21:10:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat: use unified pickle cacher & move llm config into a isolated config (#424)

Browse Source 
* simplify RDAgent conf

* add unified cacher(untested)

* fix small bugs

* fix a bug

* fix a small bug in runner

* use hash_key = None to skip cache

* fix CI

* in factor execution, ignore cache when raise exception

* add file locker to avoid mp calling

* fix CI

* use function __module__ name as folder in cache
This commit is contained in:
Xu Yang
2024-10-14 17:34:09 +08:00
committed by GitHub
parent 8341430bc9
commit 25472fc1e3
25 changed files with 299 additions and 346 deletions
Download Patch File Download Diff File Expand all files Collapse all files
rdagent/components/coder/factor_coder/CoSTEER/evaluators.py
+3 -2
View File
@@ -19,6 +19,7 @@ from rdagent.core.experiment import Task, Workspace
from rdagent.core.prompts import Prompts
from rdagent.core.utils import multiprocessing_wrapper
from rdagent.log import rdagent_logger as logger
from rdagent.oai.llm_conf import LLM_SETTINGS
from rdagent.oai.llm_utils import APIBackend
evaluate_prompts = Prompts(file_path=Path(__file__).parent.parent / "prompts.yaml")
@@ -118,7 +119,7 @@ class FactorCodeEvaluator(FactorEvaluator):
user_prompt=user_prompt,
system_prompt=system_prompt,
)
> RD_AGENT_SETTINGS.chat_token_limit
> LLM_SETTINGS.chat_token_limit
):
execution_feedback_to_render = execution_feedback_to_render[len(execution_feedback_to_render) // 2 :]
else:
@@ -521,7 +522,7 @@ class FactorFinalDecisionEvaluator(Evaluator):
user_prompt=user_prompt,
system_prompt=system_prompt,
)
> RD_AGENT_SETTINGS.chat_token_limit
> LLM_SETTINGS.chat_token_limit
):
execution_feedback_to_render = execution_feedback_to_render[len(execution_feedback_to_render) // 2 :]
else: