In [1]:
# file: pipeline/setup.py (or first notebook cell)
from __future__ import annotations
import warnings
warnings.filterwarnings(
"ignore",
message=".*str.*dtypes are inc.*",
category=UserWarning,
)
warnings.filterwarnings(
"ignore",
message=".*X does not have valid feature names.*",
category=UserWarning,
module="sklearn",
)
import json
import importlib.util
import logging
import subprocess
import sys
import time
from pathlib import Path
AUTO_INSTALL = True # set False if you never want runtime installs
_REQUIRED = {
"numpy": "numpy",
"pandas": "pandas",
"matplotlib": "matplotlib",
"sklearn": "scikit-learn",
"lightgbm": "lightgbm",
"xgboost": "xgboost",
"jinja2": "jinja2",
"joblib": "joblib",
"isort": "isort",
}
def _missing_packages(required: dict[str, str]) -> list[str]:
missing: list[str] = []
for import_name, pip_name in required.items():
if importlib.util.find_spec(import_name) is None:
missing.append(pip_name)
return missing
def ensure_dependencies(auto_install: bool = True) -> None:
"""
Ensures this kernel/interpreter has required packages.
Installs missing deps using the same interpreter as the current kernel.
"""
missing = _missing_packages(_REQUIRED)
if not missing:
return
msg = (
"Missing packages for this kernel/interpreter:\n"
f" - {', '.join(missing)}\n"
f"Kernel Python: {sys.executable}\n"
)
if not auto_install:
raise ModuleNotFoundError(
msg
+ "Install them into this interpreter, then restart the kernel."
)
result = subprocess.run(
[sys.executable, "-m", "pip", "install", "-U", *missing],
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL,
)
ensure_dependencies(AUTO_INSTALL)
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.metrics import (
roc_auc_score,
average_precision_score,
brier_score_loss,
f1_score,
precision_score,
recall_score,
roc_curve,
precision_recall_curve,
confusion_matrix,
)
from sklearn.calibration import calibration_curve, CalibratedClassifierCV
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder
from sklearn.impute import SimpleImputer
from lightgbm import LGBMClassifier
from xgboost import XGBClassifier
import joblib
logging.basicConfig(level=logging.INFO, format="%(levelname)s | %(message)s")
logger = logging.getLogger(__name__)
plt.rcParams.update(
{
"figure.facecolor": "white",
"axes.facecolor": "white",
"axes.grid": True,
"grid.alpha": 0.3,
"axes.spines.top": False,
"axes.spines.right": False,
"font.size": 11,
}
)
PALETTE = {
"lgb": "#2196F3",
"xgb": "#FF9800",
"cal": "#4CAF50",
"ref": "#9E9E9E",
"red": "#E53935",
}
print(
"Setup complete ✓",
f"(python={sys.version.split()[0]}, exe={Path(sys.executable).name}, numpy={np.__version__}, pandas={pd.__version__})",
)Setup complete ✓ (python=3.11.13, exe=python3.11, numpy=2.4.3, pandas=3.0.1)
