gavindiaz/NexQuant
1
0
Fork 0
mirror of https://github.com/NicolasBohn/NexQuant.git synced 2026-08-08 04:27:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat: a unified CoSTEER to fit more scenarios (#491)

Browse Source 
* Use ExtendedBaseSettings to replace BaseSettings

* update a more general way to pass the default setting

* update all code

* fix CI

* fix CI

* fix qlib scenario

* fix CI

* fix CI

* fix CI & add data science interfaces

* remove redundant code

* abandon costeer knowledge base v1

---------

Co-authored-by: Xu Yang <xuyang1@microsoft.com>
Co-authored-by: XianBW <36835909+XianBW@users.noreply.github.com>
This commit is contained in:
Xu Yang
2024-11-25 16:27:34 +08:00
committed by GitHub
parent a4602bdcc2
commit fce241b9f9
72 changed files with 1793 additions and 1817 deletions
Download Patch File Download Diff File Expand all files Collapse all files
rdagent/components/coder/factor_coder/prompts.yaml
+8 -42
View File
@@ -28,9 +28,9 @@ evaluator_code_feedback_v1_user: |-
{{ code }}
--------------Execution feedback:---------------
{{ execution_feedback }}
{% if factor_value_feedback is not none %}
{% if value_feedback is not none %}
--------------Factor value feedback:---------------
{{ factor_value_feedback }}
{{ value_feedback }}
{% endif %}
{% if gt_code is not none %}
--------------Ground truth Python code:---------------
@@ -62,36 +62,12 @@ evolving_strategy_factor_implementation_v1_system: |-
"code": "The Python code as a string."
}
evolving_strategy_factor_implementation_v1_user: |-
--------------Target factor information:---------------
{{ factor_information_str }}
{% if queried_similar_successful_knowledge|length != 0 %}
--------------Correct code to similar factors:---------------
{% for similar_successful_knowledge in queried_similar_successful_knowledge %}
=====Factor {{loop.index}}:=====
{{ similar_successful_knowledge.target_task.get_task_information() }}
=====Code:=====
{{ similar_successful_knowledge.implementation.code }}
{% endfor %}
{% endif %}
{% if queried_former_failed_knowledge|length != 0 %}
--------------Former failed code:---------------
{% for former_failed_knowledge in queried_former_failed_knowledge %}
=====Code to implementation {{ loop.index }}=====
{{ former_failed_knowledge.implementation.code }}
=====Feedback to implementation {{ loop.index }}=====
{{ former_failed_knowledge.feedback }}
{% endfor %}
{% endif %}
evolving_strategy_factor_implementation_v2_user: |-
--------------Target factor information:---------------
{{ factor_information_str }}
{% if queried_similar_error_knowledge|length != 0 %}
{% if not error_summary %}
{% if error_summary_critics is none %}
Recall your last failure, your implementation met some errors.
When doing other tasks, you met some similar errors but you finally solve them. Here are some examples:
{% for error_content, similar_error_knowledge in queried_similar_error_knowledge %}
@@ -108,14 +84,14 @@ evolving_strategy_factor_implementation_v2_user: |-
{{error_summary_critics}}
{% endif %}
{% endif %}
{% if queried_similar_component_knowledge|length != 0 %}
{% if queried_similar_successful_knowledge|length != 0 %}
Here are some success implements of similar component tasks, take them as references:
--------------Correct code to similar factors:---------------
{% for similar_component_knowledge in queried_similar_component_knowledge %}
{% for similar_successful_knowledge in queried_similar_successful_knowledge %}
=====Factor {{loop.index}}:=====
{{ similar_component_knowledge.target_task.get_task_information() }}
{{ similar_successful_knowledge.target_task.get_task_information() }}
=====Code:=====
{{ similar_component_knowledge.implementation.code }}
{{ similar_successful_knowledge.implementation.code }}
{% endfor %}
{% endif %}
{% if latest_attempt_to_latest_successful_execution is not none %}
@@ -189,16 +165,6 @@ select_implementable_factor_user: |-
{% endif %}
{% endfor %}
analyze_component_prompt_v1_system: |-
User is getting a new task that might consist of the components below (given in component_index: component_description):
{{all_component_content}}
You should find out what components does the new task have, and put their indices in a list.
Please response the critic in the json format. Here is an example structure for the JSON output, please strictly follow the format:
{
"component_no_list": the list containing indices of components.
}
evaluator_output_format_system: |-
User is trying to implement some factors in the following scenario:
{{ scenario }}
@@ -235,4 +201,4 @@ evaluator_final_decision_v1_user: |-
--------------Code feedback:---------------
{{ code_feedback }}
--------------Factor value feedback:---------------
{{ factor_value_feedback }}
{{ value_feedback }}