mirror of
https://github.com/NicolasBohn/NexQuant.git
synced 2026-07-28 07:57:44 +00:00
Xu Yang
1d9b4cd2ec
* use CoSTEER as component name * rename factorimplementation to avoid confusion * rename modelimplementation * align benchmark and evolving evaluators * add scenario to evaluator init function * rename all factorimplementationknowledge in CoSTEER * remove all scenario related information in component * remove useless code --------- Co-authored-by: xuyang1 <xuyang1@microsoft.com>
52 lines
1.7 KiB
Python
52 lines
1.7 KiB
Python
from pathlib import Path
|
|
|
|
import pandas as pd
|
|
|
|
# render it with jinja
|
|
from jinja2 import Environment, StrictUndefined
|
|
|
|
from rdagent.components.coder.factor_coder.config import FACTOR_IMPLEMENT_SETTINGS
|
|
|
|
TPL = """
|
|
{{file_name}}
|
|
```{{type_desc}}
|
|
{{content}}
|
|
````
|
|
"""
|
|
# Create a Jinja template from the string
|
|
JJ_TPL = Environment(undefined=StrictUndefined).from_string(TPL)
|
|
|
|
|
|
def get_data_folder_intro():
|
|
"""Directly get the info of the data folder.
|
|
It is for preparing prompting message.
|
|
"""
|
|
content_l = []
|
|
for p in Path(FACTOR_IMPLEMENT_SETTINGS.file_based_execution_data_folder).iterdir():
|
|
if p.name.endswith(".h5"):
|
|
df = pd.read_hdf(p)
|
|
# get df.head() as string with full width
|
|
pd.set_option("display.max_columns", None) # or 1000
|
|
pd.set_option("display.max_rows", None) # or 1000
|
|
pd.set_option("display.max_colwidth", None) # or 199
|
|
rendered = JJ_TPL.render(
|
|
file_name=p.name,
|
|
type_desc="generated by `pd.read_hdf(filename).head()`",
|
|
content=df.head().to_string(),
|
|
)
|
|
content_l.append(rendered)
|
|
elif p.name.endswith(".md"):
|
|
with open(p) as f:
|
|
content = f.read()
|
|
rendered = JJ_TPL.render(
|
|
file_name=p.name,
|
|
type_desc="markdown",
|
|
content=content,
|
|
)
|
|
content_l.append(rendered)
|
|
else:
|
|
raise NotImplementedError(
|
|
f"file type {p.name} is not supported. Please implement its description function.",
|
|
)
|
|
return "\n ----------------- file splitter -------------\n".join(content_l)
|