modifications documentation

Author: LacombeLouis

doc/tutorial_regression.rst

@@ -577,7 +577,6 @@ next figure.
     fig = plt.figure()
     heteroscedastic_coverage.T.plot.bar(figsize=(12, 4), alpha=0.7)
     plt.axhline(0.95, ls="--", color="k")
-    plt.ylim([0.8, 1])
     plt.ylabel("Conditional coverage")
     plt.xlabel("x bins")
     plt.xticks(rotation=0)

doc/tutorial_regression_files/tutorial_regression_73_0.png

@@ -15,7 +15,6 @@
 import scipy
 import numpy as np
 from sklearn.linear_model import LinearRegression, QuantileRegressor
-from sklearn.model_selection import train_test_split
 from sklearn.pipeline import Pipeline
 from sklearn.preprocessing import PolynomialFeatures
 from matplotlib import pyplot as plt
@@ -25,6 +24,8 @@
 from mapie.subsample import Subsample
 from mapie._typing import NDArray
 
+random_state = 42
+
 
 def f(x: NDArray) -> NDArray:
     """Polynomial function used to generate one-dimensional data"""
@@ -59,7 +60,7 @@ def get_heteroscedastic_data(
         [3]: y_true
         [4]: y_true_sigma
     """
-    np.random.seed(59)
+    np.random.seed(random_state)
     q95 = scipy.stats.norm.ppf(0.95)
     X_train = np.linspace(0, 1, n_train)
     X_true = np.linspace(0, 1, n_true)
@@ -158,22 +159,8 @@ def plot_1d_data(
             polyn_model_quant,
             **params
         )
-        X_train_split, X_calib, y_train_spit, y_calib = train_test_split(
-            X_train,
-            y_train,
-            test_size=0.3,
-            random_state=1
-        )
-        mapie.fit(
-            X_train_split.reshape(-1, 1),
-            y_train_spit,
-            X_calib=X_calib.reshape(-1, 1),
-            y_calib=y_calib
-        )
-        y_pred, y_pis = mapie.predict(
-            X_test.reshape(-1, 1)
-        )
-        X_train, y_train = X_train_split, y_train_spit
+        mapie.fit(X_train.reshape(-1, 1), y_train, random_state=random_state)
+        y_pred, y_pis = mapie.predict(X_test.reshape(-1, 1))
     else:
         mapie = MapieRegressor(  # type: ignore
             polyn_model,

doc/tutorial_regression_files/tutorial_regression_74_0.png

@@ -13,7 +13,6 @@
 import scipy
 import numpy as np
 from sklearn.linear_model import LinearRegression, QuantileRegressor
-from sklearn.model_selection import train_test_split
 from sklearn.pipeline import Pipeline
 from sklearn.preprocessing import PolynomialFeatures
 from matplotlib import pyplot as plt
@@ -23,6 +22,8 @@
 from mapie.subsample import Subsample
 from mapie._typing import NDArray
 
+random_state = 42
+
 
 def f(x: NDArray) -> NDArray:
     """Polynomial function used to generate one-dimensional data"""
@@ -57,7 +58,7 @@ def get_homoscedastic_data(
         [3]: y_true
         [4]: y_true_sigma
     """
-    np.random.seed(59)
+    np.random.seed(random_state)
     q95 = scipy.stats.norm.ppf(0.95)
     X_train = np.linspace(0, 1, n_train)
     X_true = np.linspace(0, 1, n_true)
@@ -156,18 +157,7 @@ def plot_1d_data(
             polyn_model_quant,
             **params
         )
-        X_train, X_calib, y_train, y_calib = train_test_split(
-            X_train,
-            y_train,
-            test_size=0.3,
-            random_state=1
-        )
-        mapie.fit(
-            X_train.reshape(-1, 1),
-            y_train,
-            X_calib=X_calib.reshape(-1, 1),
-            y_calib=y_calib
-        )
+        mapie.fit(X_train.reshape(-1, 1), y_train, random_state=random_state)
         y_pred, y_pis = mapie.predict(X_test.reshape(-1, 1))
     else:
         mapie = MapieRegressor(  # type: ignore

examples/regression/1-quickstart/plot_heteroscedastic_1d_data.py

@@ -11,7 +11,6 @@
 
 import numpy as np
 from sklearn.linear_model import LinearRegression, QuantileRegressor
-from sklearn.model_selection import train_test_split
 from sklearn.pipeline import Pipeline
 from sklearn.preprocessing import PolynomialFeatures
 import matplotlib.pyplot as plt
@@ -22,6 +21,7 @@
 from mapie._typing import NDArray
 
 F = TypeVar("F", bound=Callable[..., Any])
+random_state = 42
 
 
 # Functions for generating our dataset
@@ -60,7 +60,7 @@ def get_1d_data_with_normal_distrib(
         [3]: y_test
         [4]: y_mesh
     """
-    np.random.seed(42)
+    np.random.seed(random_state)
     X_train = np.random.normal(mu, sigma, n_samples)
     X_test = np.arange(mu - 4 * sigma, mu + 4 * sigma, sigma / 20.0)
     y_train, y_mesh, y_test = funct(X_train), funct(X_test), funct(X_test)
@@ -114,21 +114,8 @@ def get_1d_data_with_normal_distrib(
             polyn_model_quant,
             **params
         )
-        X_train, X_calib, y_train, y_calib = train_test_split(
-            X_train,
-            y_train,
-            test_size=0.3,
-            random_state=1
-        )
-        mapie.fit(
-            X_train,
-            y_train,
-            X_calib=X_calib,
-            y_calib=y_calib
-        )
-        y_pred[strategy], y_pis[strategy] = mapie.predict(
-            X_test
-        )
+        mapie.fit(X_train, y_train, random_state=random_state)
+        y_pred[strategy], y_pis[strategy] = mapie.predict(X_test)
     else:
         mapie = MapieRegressor(polyn_model, **params)  # type: ignore
         mapie.fit(X_train, y_train)

examples/regression/1-quickstart/plot_homoscedastic_1d_data.py

@@ -429,7 +429,10 @@ def test_calib_dataset_is_none_with_sample_weight() -> None:
 
 
 def test_calib_dataset_is_none_vs_defined() -> None:
-    """Test that the fit method works when X_calib or y_calib is None."""
+    """
+    Test that for the same results whether you split before
+    or in the fit method.
+    """
     mapie = MapieQuantileRegressor()
     mapie_defined = clone(mapie)
     mapie.fit(X, y, calib_size=0.5, random_state=random_state)

examples/regression/2-advanced-analysis/plot_both_uncertainties.py

@@ -456,7 +456,6 @@ heteroscedastic_coverage = get_heteroscedastic_coverage(y_test, y_pis, STRATEGIE
 fig = plt.figure()
 heteroscedastic_coverage.T.plot.bar(figsize=(12, 4), alpha=0.7)
 plt.axhline(0.95, ls="--", color="k")
-plt.ylim([0.8, 1])
 plt.ylabel("Conditional coverage")
 plt.xlabel("x bins")
 plt.xticks(rotation=0)