added benchmark and reference to this example in the docs

benchmarks/cb/decomposition.py

@@ -0,0 +1,17 @@
+# flake8: noqa
+import heat as ht
+from mpi4py import MPI
+from perun import monitor
+from heat.decomposition import IncrementalPCA
+
+
+@monitor()
+def incremental_pca_split0(list_of_X, n_components):
+    ipca = IncrementalPCA(n_components=n_components)
+    for X in list_of_X:
+        ipca.partial_fit(X)
+
+
+def run_decomposition_benchmarks():
+    list_of_X = [ht.random.rand(50000, 500, split=0) for _ in range(10)]
+    incremental_pca_split0(list_of_X, 50)

benchmarks/cb/main.py

@@ -10,8 +10,10 @@
 from cluster import run_cluster_benchmarks
 from manipulations import run_manipulation_benchmarks
 from preprocessing import run_preprocessing_benchmarks
+from decomposition import run_decomposition_benchmarks
 
 run_linalg_benchmarks()
 run_cluster_benchmarks()
 run_manipulation_benchmarks()
 run_preprocessing_benchmarks()
+run_decomposition_benchmarks()

heat/decomposition/pca.py

@@ -286,6 +286,8 @@ class IncrementalPCA(ht.TransformMixin, ht.BaseEstimator):
 
     This class allows for incremental updates of the PCA model. This is especially useful for large data sets that do not fit into memory.
 
+    An example how to apply this class is given in, e.g., `benchmarks/cb/decomposition.py`.
+
     Parameters
     ----------
     n_components : int, optional