Cherry pick commits from PRs 1088, 1100, 1120, 1140, and 1145 for documentation updates (#1153)

Author: nvkevlu

docs/examples/xgboost.rst

@@ -0,0 +1,27 @@
+Federated XGBoost
+=================
+
+
+Overview
+--------
+
+NVFlare supports federated learning using popular gradient boosting library XGBoost.
+It uses XGBoost library with federated plugin (xgboost version >= 1.7.0rc1) to perform the learning.
+
+Using XGBoost with NVFlare has following benefits compared with running federated XGBoost directly,
+
+* XGBoost instance's life-cycle is managed by NVFlare. Both XGBoost client and server
+  are started/stopped automatically by NVFlare workflow.
+* For histogram-based XGBoost federated server can be configured automatically with auto-assigned port number.
+* When mutual TLS is used, the certificates are managed by NVFlare using existing
+  provisioning process.
+* No need to manually configure each instance. Instance specific parameters
+  like code:`rank` are assigned automatically by the NVFlare controller.
+
+Examples
+--------
+
+Basic components to run XGBoost are already included with NVFlare distribution.
+Most XGBoost jobs can be created without custom code.
+
+Please refer to :code:`NVFlare/examples/xgboost` for more details.

docs/index.rst

@@ -33,7 +33,7 @@ NVIDIA FLARE is built on a componentized architecture that gives you the flexibi
 
 Getting Started
 ===============
-For first-time users and FL researchers, FLARE provides the FL Simulator that allows you to build, test, and deploy applications locally.  The :ref:`Getting Started guide <getting_started>` covers installation and walks through an example application using the FL Simulator.
+For first-time users and FL researchers, FLARE provides the :ref:`fl_simulator` that allows you to build, test, and deploy applications locally.  The :ref:`Getting Started guide <getting_started>` covers installation and walks through an example application using the FL Simulator.
 
 When you are ready to for a secure, distributed deployment, the :ref:`Real World Federated Learning <real_world_fl>` section covers the tools and process required to deploy and operate a secure, real-world FLARE project.
 

docs/programming_guide/provisioning_system.rst

@@ -283,6 +283,67 @@ A new builder to write 'gateway.conf' can be implemented as follows (for referen
                   f.write(f"name = {gw.name}\n")
                   f.write(f"port = {port}\n")
 
+.. _distribution_builder:
+
+Case 4: adding a builder for enabling the creation of zip archives for the startup kits
+---------------------------------------------------------------------------------------
+DistributionBuilder was included in NVIDIA FLARE before version 2.2.1 but has been removed from the
+default builders. You can make this builder available and add it as a builder in project.yml if you want to zip the startup kits::
+
+    import os
+    import shutil
+    import subprocess
+
+    from nvflare.lighter.spec import Builder, Project
+    from nvflare.lighter.utils import generate_password
+
+    class DistributionBuilder(Builder):
+        def __init__(self, zip_password=False):
+            """Build the zip files for each folder.
+            Creates the zip files containing the archives for each startup kit. It will add password protection if the
+            argument (zip_password) is true.
+            Args:
+                zip_password: if true, will create zipped packages with passwords
+            """
+            self.zip_password = zip_password
+
+        def build(self, project: Project, ctx: dict):
+            """Create a zip for each individual folder.
+            Note that if zip_password is True, the zip command will be used to encrypt zip files.  Users have to
+            install this zip utility before provisioning.  In Ubuntu system, use this command to install zip utility:
+            sudo apt-get install zip
+            Args:
+                project (Project): project instance
+                ctx (dict): the provision context
+            """
+            wip_dir = self.get_wip_dir(ctx)
+            dirs = [
+                name
+                for name in os.listdir(wip_dir)
+                if os.path.isdir(os.path.join(wip_dir, name)) and "nvflare_" not in name
+            ]
+            for dir in dirs:
+                dest_zip_file = os.path.join(wip_dir, f"{dir}")
+                if self.zip_password:
+                    pw = generate_password()
+                    run_args = ["zip", "-rq", "-P", pw, dest_zip_file + ".zip", ".", "-i", "startup/*"]
+                    os.chdir(dest_zip_file)
+                    try:
+                        subprocess.run(run_args)
+                        print(f"Password {pw} on {dir}.zip")
+                    except FileNotFoundError:
+                        raise RuntimeError("Unable to zip folders with password.  Maybe the zip utility is not installed.")
+                    finally:
+                        os.chdir(os.path.join(dest_zip_file, ".."))
+                else:
+                    shutil.make_archive(dest_zip_file, "zip", root_dir=os.path.join(wip_dir, dir), base_dir="startup")
+
+If the above code is made available at ``nvflare.lighter.impl.workspace.DistributionBuilder``, add the following to your project.yml at the bottom of the list of builders::
+
+    path: nvflare.lighter.impl.workspace.DistributionBuilder
+    args:
+      zip_password: true
+
 Takeaways for Custom Builders
 -----------------------------
 From the cases shown previously, implementing your own Builders only requires the following steps:

docs/publications_and_talks.md

@@ -40,6 +40,7 @@ NVIDIA FLARE related blogs and other media.
 
 #### 2019
 * **2019-12** [Federated Learning powered by NVIDIA Clara](https://developer.nvidia.com/blog/federated-learning-clara/) (NVIDIA Technical Blog)
+* **2019-10** [What is federated learning - in Chinese](https://blogs.nvidia.com.tw/2019/10/13/what-is-federated-learning/) (NVIDIA Technical Blog)
 * **2019-10** [NVIDIA Research: First Privacy-Preserving Federated Learning System for Medical Imaging](https://www.youtube.com/watch?v=Jy7ozgwovgg) (NVIDIA video)
 
 ## Talks
@@ -58,3 +59,6 @@ Recent talks and Webinars covering federated learning research and NVIDIA FLARE.
 * **2021-09** [Federated Learning](https://www.youtube.com/watch?v=YeYO4JGTBb0&amp) ([MONAI MICCAI Bootcamp 2021](https://www.gpuhackathons.org/event/monai-miccai-bootcamp-2021))
 * **2021-03** [NVIDIA FLARE: An Open Federated Learning Platform](https://www.nvidia.com/en-us/on-demand/session/gtcspring22-se1991/) ([GTC Spring 2022](https://www.nvidia.com/gtc/))
 * **2021-03** [Federated Learning for Healthcare – Collaborative AI without Sharing Patient Data ](https://www.youtube.com/watch?v=xr_eJp3ctzw) ([Data Science Seminar](https://www.dkfz.de/en/datascience/seminar/Rieke.html))
+
+#### 2020
+* **2020-11** [Federated Learning for Medical Imaging - in Chinese](https://www.youtube.com/watch?v=CiPdALrNEjU) (NVIDIA Taiwan)

docs/quickstart.rst

@@ -0,0 +1,5 @@
+##########
+Quickstart
+##########
+
+See :ref:`quickstart`.

docs/real_world_fl.rst

@@ -8,7 +8,9 @@ This section shows how to use NVIDIA FLARE to deploy and operate an FL system.
 A reference application will be used here to show provisioning and basic operation
 of the system through the admin client. You will find information here about setting up the
 system with all the available components and the Admin API for
-operating FL. For more details on what you can do with apps with custom components and
+operating FL. For instructions on how to set up the :ref:`nvflare_dashboard_ui` added in 2.2.1 to
+help gather information to provision a project and distribute startup kits, see :ref:`dashboard_api`.
+For more details on what you can do with apps with custom components and
 the flexibility that the Controller and Worker APIs bring, see the :ref:`programming_guide`.
 
 You can also see some `example applications <https://github.com/NVIDIA/NVFlare/tree/main/examples>`_ integrating with
@@ -24,5 +26,4 @@ to see the capabilities of the system and how it can be operated.
    real_world_fl/application
    real_world_fl/job
    real_world_fl/workspace
-   real_world_fl/authorization
-
+   user_guide/federated_authorization

docs/real_world_fl/application.rst

@@ -107,16 +107,15 @@ the client config should have the following in order to configure it as an Execu
 
     Configuration of Executor Tasks is ignored here.
 
-Please follow :ref:`quickstart:Quickstart` to learn more.
+Please follow :ref:`quickstart` to learn more.
 
 .. _troubleshooting_byoc:
 
 Troubleshooting BYOC
 ====================
-There is an ``enable_byoc`` flag for each participant that can be set at provisioning, and if that is disabled, even if
-you have custom code in your application folder, it will not be loaded. There is also a setting for ``allow_byoc``
-through the authorization rule groups. This controls whether or not apps containing BYOC code will be allowed to be
-uploaded and deployed.
+In 2.2.1, authorization has been redesigned and BYOC is no longer controlled through settings at provisioning, but
+instead by each site's authorization.json (in the local folder of the workspace). BYOC is a right and can be restricted
+to certain roles or even orgs or users. See :ref:`federated_authorization` for details.
 
 *********
 Resources