Pushing the docs to dev/ for branch: main, commit 13d56e9c90e74efbcfc7a3281a74f734fb3d2272

Author: sklearn-ci

dev/.buildinfo

@@ -1,4 +1,4 @@
 # Sphinx build info version 1
 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
-config: 8a3c7950512bce9aaa2a3a53e2716be7
+config: 33c62d0b3c65f9c70b9234643a63348f
 tags: 645f666f9bcd5a90fca523b33c5a78b7

dev/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip

@@ -386,7 +386,7 @@
       },
       "outputs": [],
       "source": [
-        "import numpy as np\n\n# unused but required import for doing 3d projections with matplotlib < 3.2\nimport mpl_toolkits.mplot3d  # noqa: F401\n\nfrom sklearn.inspection import partial_dependence\n\nfig = plt.figure(figsize=(5.5, 5))\n\nfeatures = (\"temp\", \"humidity\")\npdp = partial_dependence(\n    hgbdt_model, X_train, features=features, kind=\"average\", grid_resolution=10\n)\nXX, YY = np.meshgrid(pdp[\"values\"][0], pdp[\"values\"][1])\nZ = pdp.average[0].T\nax = fig.add_subplot(projection=\"3d\")\nfig.add_axes(ax)\n\nsurf = ax.plot_surface(XX, YY, Z, rstride=1, cstride=1, cmap=plt.cm.BuPu, edgecolor=\"k\")\nax.set_xlabel(features[0])\nax.set_ylabel(features[1])\nfig.suptitle(\n    \"PD of number of bike rentals on\\nthe temperature and humidity GBDT model\",\n    fontsize=16,\n)\n# pretty init view\nax.view_init(elev=22, azim=122)\nclb = plt.colorbar(surf, pad=0.08, shrink=0.6, aspect=10)\nclb.ax.set_title(\"Partial\\ndependence\")\nplt.show()"
+        "import numpy as np\n\n# unused but required import for doing 3d projections with matplotlib < 3.2\nimport mpl_toolkits.mplot3d  # noqa: F401\n\nfrom sklearn.inspection import partial_dependence\n\nfig = plt.figure(figsize=(5.5, 5))\n\nfeatures = (\"temp\", \"humidity\")\npdp = partial_dependence(\n    hgbdt_model, X_train, features=features, kind=\"average\", grid_resolution=10\n)\nXX, YY = np.meshgrid(pdp[\"grid_values\"][0], pdp[\"grid_values\"][1])\nZ = pdp.average[0].T\nax = fig.add_subplot(projection=\"3d\")\nfig.add_axes(ax)\n\nsurf = ax.plot_surface(XX, YY, Z, rstride=1, cstride=1, cmap=plt.cm.BuPu, edgecolor=\"k\")\nax.set_xlabel(features[0])\nax.set_ylabel(features[1])\nfig.suptitle(\n    \"PD of number of bike rentals on\\nthe temperature and humidity GBDT model\",\n    fontsize=16,\n)\n# pretty init view\nax.view_init(elev=22, azim=122)\nclb = plt.colorbar(surf, pad=0.08, shrink=0.6, aspect=10)\nclb.ax.set_title(\"Partial\\ndependence\")\nplt.show()"
       ]
     }
   ],

dev/_downloads/21b82d82985712b5de6347f382c77c86/plot_partial_dependence.ipynb

@@ -704,7 +704,7 @@
       },
       "outputs": [],
       "source": [
-        "fig, axes = plt.subplots(ncols=3, figsize=(12, 4), sharey=True)\nfig.suptitle(\"Non-linear regression models\")\npredictions = [\n    one_hot_poly_predictions,\n    cyclic_spline_poly_predictions,\n    gbrt_predictions,\n]\nlabels = [\n    \"One hot + polynomial kernel\",\n    \"Splines + polynomial kernel\",\n    \"Gradient Boosted Trees\",\n]\nfor ax, pred, label in zip(axes, predictions, labels):\n    ax.scatter(y.iloc[test_0].values, pred, alpha=0.3, label=label)\n    ax.plot([0, 1], [0, 1], \"--\", label=\"Perfect model\")\n    ax.set(\n        xlim=(0, 1),\n        ylim=(0, 1),\n        xlabel=\"True demand\",\n        ylabel=\"Predicted demand\",\n    )\n    ax.legend()\n\nplt.show()"
+        "from sklearn.metrics import PredictionErrorDisplay\n\nfig, axes = plt.subplots(nrows=2, ncols=3, figsize=(13, 7), sharex=True, sharey=\"row\")\nfig.suptitle(\"Non-linear regression models\", y=1.0)\npredictions = [\n    one_hot_poly_predictions,\n    cyclic_spline_poly_predictions,\n    gbrt_predictions,\n]\nlabels = [\n    \"One hot +\\npolynomial kernel\",\n    \"Splines +\\npolynomial kernel\",\n    \"Gradient Boosted\\nTrees\",\n]\nplot_kinds = [\"actual_vs_predicted\", \"residual_vs_predicted\"]\nfor axis_idx, kind in enumerate(plot_kinds):\n    for ax, pred, label in zip(axes[axis_idx], predictions, labels):\n        disp = PredictionErrorDisplay.from_predictions(\n            y_true=y.iloc[test_0],\n            y_pred=pred,\n            kind=kind,\n            scatter_kwargs={\"alpha\": 0.3},\n            ax=ax,\n        )\n        ax.set_xticks(np.linspace(0, 1, num=5))\n        if axis_idx == 0:\n            ax.set_yticks(np.linspace(0, 1, num=5))\n            ax.legend(\n                [\"Best model\", label],\n                loc=\"upper center\",\n                bbox_to_anchor=(0.5, 1.3),\n                ncol=2,\n            )\n        ax.set_aspect(\"equal\", adjustable=\"box\")\nplt.show()"
       ]
     },
     {

dev/_downloads/21fba8cadc21699d2b4699b4ccdad10f/plot_beta_divergence.ipynb

@@ -774,29 +774,40 @@ def periodic_spline_transformer(period, n_splines=None, degree=3):
 #
 # Let us finally get a more quantative look at the prediction errors of those
 # three models using the true vs predicted demand scatter plots:
-fig, axes = plt.subplots(ncols=3, figsize=(12, 4), sharey=True)
-fig.suptitle("Non-linear regression models")
+from sklearn.metrics import PredictionErrorDisplay
+
+fig, axes = plt.subplots(nrows=2, ncols=3, figsize=(13, 7), sharex=True, sharey="row")
+fig.suptitle("Non-linear regression models", y=1.0)
 predictions = [
     one_hot_poly_predictions,
     cyclic_spline_poly_predictions,
     gbrt_predictions,
 ]
 labels = [
-    "One hot + polynomial kernel",
-    "Splines + polynomial kernel",
-    "Gradient Boosted Trees",
+    "One hot +\npolynomial kernel",
+    "Splines +\npolynomial kernel",
+    "Gradient Boosted\nTrees",
 ]
-for ax, pred, label in zip(axes, predictions, labels):
-    ax.scatter(y.iloc[test_0].values, pred, alpha=0.3, label=label)
-    ax.plot([0, 1], [0, 1], "--", label="Perfect model")
-    ax.set(
-        xlim=(0, 1),
-        ylim=(0, 1),
-        xlabel="True demand",
-        ylabel="Predicted demand",
-    )
-    ax.legend()
-
+plot_kinds = ["actual_vs_predicted", "residual_vs_predicted"]
+for axis_idx, kind in enumerate(plot_kinds):
+    for ax, pred, label in zip(axes[axis_idx], predictions, labels):
+        disp = PredictionErrorDisplay.from_predictions(
+            y_true=y.iloc[test_0],
+            y_pred=pred,
+            kind=kind,
+            scatter_kwargs={"alpha": 0.3},
+            ax=ax,
+        )
+        ax.set_xticks(np.linspace(0, 1, num=5))
+        if axis_idx == 0:
+            ax.set_yticks(np.linspace(0, 1, num=5))
+            ax.legend(
+                ["Best model", label],
+                loc="upper center",
+                bbox_to_anchor=(0.5, 1.3),
+                ncol=2,
+            )
+        ax.set_aspect("equal", adjustable="box")
 plt.show()
 # %%
 # This visualization confirms the conclusions we draw on the previous plot.

dev/_downloads/6d09465eed1ee4ede505244049097627/plot_beta_divergence.py

@@ -532,7 +532,7 @@
 pdp = partial_dependence(
     hgbdt_model, X_train, features=features, kind="average", grid_resolution=10
 )
-XX, YY = np.meshgrid(pdp["values"][0], pdp["values"][1])
+XX, YY = np.meshgrid(pdp["grid_values"][0], pdp["grid_values"][1])
 Z = pdp.average[0].T
 ax = fig.add_subplot(projection="3d")
 fig.add_axes(ax)