eddiebergman: Merge HOTFIX master 0.14.3 into dev

Author: Github Actions

development/.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: 5cb66ffaf1a93d48276b86c4c7778f5e
+config: 1281a8de7e72cc26a3b0fce1416780a7
 tags: 645f666f9bcd5a90fca523b33c5a78b7

development/_downloads/6507de4916ad481ba20cb60e6dc3309e/example_parallel_manual_spawning_python.py

@@ -11,11 +11,17 @@
 This example shows how to start the dask scheduler and spawn
 workers for *Auto-sklearn* manually from the command line. Use this example
 as a starting point to parallelize *Auto-sklearn* across multiple
-machines. If you want to start everything manually from within Python
-please see :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`.
+machines.
+
 To run *Auto-sklearn* in parallel on a single machine check out the example
 :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.
 
+If you want to start everything manually from within Python
+please see ``:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py``.
+
+**NOTE:** Above example is disabled due to issue https://github.com/dask/distributed/issues/5627
+
+
 You can learn more about the dask command line interface from
 https://docs.dask.org/en/latest/setup/cli.html.
 

development/_downloads/986156f70b0b7eee4d669b791b35bd6d/example_parallel_manual_spawning_python.ipynb

@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Parallel Usage: Spawning workers from the command line\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually from the command line. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines. If you want to start everything manually from within Python\nplease see `sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`.\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nYou can learn more about the dask command line interface from\nhttps://docs.dask.org/en/latest/setup/cli.html.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
+        "\n# Parallel Usage: Spawning workers from the command line\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually from the command line. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines.\n\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nIf you want to start everything manually from within Python\nplease see ``:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py``.\n\n**NOTE:** Above example is disabled due to issue https://github.com/dask/distributed/issues/5627\n\n\nYou can learn more about the dask command line interface from\nhttps://docs.dask.org/en/latest/setup/cli.html.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
       ]
     },
     {

development/_downloads/baf53fc945368668a0cd202acebc6220/example_parallel_manual_spawning_cli.py

@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.estimators &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.estimators &#8212; AutoSklearn 0.14.3 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.3</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">