Pushing the docs to dev/ for branch: master, commit ed5a0cac6b0bdf2ff9920a7565efa0bff12f485c

Author: sklearn-ci

dev/_downloads/auto_examples_jupyter.zip

@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Empirical evaluation of the impact of k-means initialization\n\n\nEvaluate the ability of k-means initializations strategies to make\nthe algorithm convergence robust as measured by the relative standard\ndeviation of the inertia of the clustering (i.e. the sum of distances\nto the nearest cluster center).\n\nThe first plot shows the best inertia reached for each combination\nof the model (``KMeans`` or ``MiniBatchKMeans``) and the init method\n(``init=\"random\"`` or ``init=\"kmeans++\"``) for increasing values of the\n``n_init`` parameter that controls the number of initializations.\n\nThe second plot demonstrate one single run of the ``MiniBatchKMeans``\nestimator using a ``init=\"random\"`` and ``n_init=1``. This run leads to\na bad convergence (local optimum) with estimated centers stuck\nbetween ground truth clusters.\n\nThe dataset used for evaluation is a 2D grid of isotropic Gaussian\nclusters widely spaced.\n\n"
+        "\n# Empirical evaluation of the impact of k-means initialization\n\n\nEvaluate the ability of k-means initializations strategies to make\nthe algorithm convergence robust as measured by the relative standard\ndeviation of the inertia of the clustering (i.e. the sum of squared\ndistances to the nearest cluster center).\n\nThe first plot shows the best inertia reached for each combination\nof the model (``KMeans`` or ``MiniBatchKMeans``) and the init method\n(``init=\"random\"`` or ``init=\"kmeans++\"``) for increasing values of the\n``n_init`` parameter that controls the number of initializations.\n\nThe second plot demonstrate one single run of the ``MiniBatchKMeans``\nestimator using a ``init=\"random\"`` and ``n_init=1``. This run leads to\na bad convergence (local optimum) with estimated centers stuck\nbetween ground truth clusters.\n\nThe dataset used for evaluation is a 2D grid of isotropic Gaussian\nclusters widely spaced.\n\n"
       ]
     },
     {

dev/_downloads/auto_examples_python.zip

@@ -5,8 +5,8 @@
 
 Evaluate the ability of k-means initializations strategies to make
 the algorithm convergence robust as measured by the relative standard
-deviation of the inertia of the clustering (i.e. the sum of distances
-to the nearest cluster center).
+deviation of the inertia of the clustering (i.e. the sum of squared
+distances to the nearest cluster center).
 
 The first plot shows the best inertia reached for each combination
 of the model (``KMeans`` or ``MiniBatchKMeans``) and the init method