From c6772d64cabbecbb2fb0614142a2dd6f4e81523d Mon Sep 17 00:00:00 2001
From: jawadhussein462 <jawadhussein462@gmail.com>
Date: Thu, 19 Dec 2024 16:08:14 +0100
Subject: [PATCH] EHC: V1 CQR improve docstrings

---
 mapie_v1/regression.py | 56 +++++++++++++++++++++---------------------
 1 file changed, 28 insertions(+), 28 deletions(-)

diff --git a/mapie_v1/regression.py b/mapie_v1/regression.py
index 4e2de7844..62967e8f6 100644
--- a/mapie_v1/regression.py
+++ b/mapie_v1/regression.py
@@ -834,43 +834,43 @@ def predict(
 
 class ConformalizedQuantileRegressor:
     """
-    A conformal quantile regression model that generates prediction intervals
-    using quantile regression as the base estimator.
+    A model that combines quantile regression with conformal prediction to
+    generate reliable prediction intervals with specified coverage levels.
 
-    This approach provides prediction intervals by leveraging
-    quantile predictions and applying conformal adjustments to ensure coverage.
+    The `ConformalizedQuantileRegressor` leverages quantile regression as its
+    base estimator to predict conditional quantiles of the target variable,
+    and applies conformal adjustments to ensure prediction intervals achieve
+    the desired confidence levels. This approach is particularly useful in
+    uncertainty quantification for regression tasks.
 
     Parameters
     ----------
-    estimator : RegressorMixin, default=QuantileRegressor()
-        The base quantile regression estimator used to generate point and
-        interval predictions.
-
-    confidence_level : Union[float, List[float]], default=0.9
+    estimator : Union[RegressorMixin, Pipeline,
+                    List[Union[RegressorMixin, Pipeline]]]
+        The base quantile regression model(s) used to estimate the target
+        quantiles.
+        - In `fit` mode (`prefit=False`), this should be a single quantile
+          regression estimator (e.g., `QuantileRegressor`) or a pipeline
+          combining preprocessing and regression.
+        - In `prefit` mode (`prefit=True`), this must be a list of three
+          fitted quantile regression estimators, corresponding to the lower,
+          upper, and median quantiles. These models are expected to be
+          pre-trained and aligned with the target quantiles.
+
+    confidence_level : float default=0.9
         The confidence level(s) for the prediction intervals, indicating the
-        desired coverage probability of the prediction intervals. If a float
-        is provided, it represents a single confidence level. If a list,
-        multiple prediction intervals for each specified confidence level
-        are returned.
-
-    conformity_score : Union[str, BaseRegressionScore], default="absolute"
-        The conformity score method used to calculate the conformity error.
-        Valid options: TODO : reference here the valid options, once the list
-        has been be created during the implementation.
-        See: TODO : reference conformity score classes or documentation
-
-        A custom score function inheriting from BaseRegressionScore may also
-        be provided.
+        desired coverage probability of the prediction intervals.
 
-    random_state : Optional[Union[int, np.random.RandomState]], default=None
-        A seed or random state instance to ensure reproducibility in any random
-        operations within the regressor.
+    prefit : bool, default=False
+        If `True`, assumes the base estimators are already fitted.
+        When set to `True`, the `fit` method cannot be called and the
+        provided estimators should be pre-trained.
 
     Methods
     -------
     fit(X_train, y_train, fit_params=None) -> Self
-        Fits the base estimator to the training data and initializes internal
-        parameters required for conformal prediction.
+        Trains the base quantile regression estimator on the provided data.
+        Not applicable if `prefit=True`.
 
     conformalize(X_conf, y_conf, predict_params=None) -> Self
         Calibrates the model on provided data, adjusting the prediction
@@ -987,7 +987,7 @@ def conformalize(
         """
         Calibrates the model on the provided data, adjusting the prediction
         intervals based on quantile predictions and specified confidence
-        levels. This step analyzes the conformity scores and refines the
+        level. This step analyzes the conformity scores and refines the
         intervals to ensure desired coverage.
 
         Parameters