PredictionCalibration

PredictionCalibration

The function predicts the probability based on the calibration model.

Parameters:

• x_data: pd.DataFrame
  A series of predictions of the raw models.
• model: LogisticRegression
  A Logistic Regression Model for calibration.
• logprob: bool
  A boolean variable whether to calculate a logarithm of probabilities.

Returns:

• pred: np.ndarray
  A NumPy array of the calibrated probabilities.

Exceptions:

• TypeError:
  Raised if x_data parameter is not a pandas DataFrame object.
  Raised if logprob parameter is not logical.
  

Example:

import pandas as pd
from sklearn.linear_model import LogisticRegression
from combat.calibration import PredictionCalibration, CalibrationModel

# Sample input data
x_data = pd.DataFrame({'A': [1, 2, 3, 4, 5], 'B': [6, 7, 8, 9, 10]})
model = CalibrationModel(*args)
logprob = False

# Predict calibrated probabilities
pred = PredictionCalibration(x_data, model, logprob)
print(pred)