update fl docs (Project-MONAI#5364)

Signed-off-by: Holger Roth <hroth@nvidia.com> Fixes Project-MONAI#5363. ### Description Update what's new and module description with new Monai FL features. ### Types of changes <!--- Put an `x` in all the boxes that apply, and remove the not applicable items --> - [x] Non-breaking change (fix or new feature that would not break existing functionality). - [ ] Breaking change (fix or new feature that would cause existing functionality to change). - [ ] New tests added to cover the changes. - [ ] Integration tests passed locally by running `./runtests.sh -f -u --net --coverage`. - [ ] Quick tests passed locally by running `./runtests.sh --quick --unittests --disttests`. - [ ] In-line docstrings updated. - [x] Documentation updated, tested `make html` command in the `docs/` folder. Signed-off-by: Holger Roth <hroth@nvidia.com>

Author: holgerroth

docs/source/fl.rst

@@ -6,11 +6,23 @@ Federated Learning
 ==================
 .. currentmodule:: monai.fl.client
 
-`ClientAlgo`
-------------
+`Client Base Classes`
+---------------------
+
+.. autoclass:: BaseClient
+ :members:
 
 .. autoclass:: ClientAlgo
  :members:
 
+.. autoclass:: ClientAlgoStats
+ :members:
+
+`MONAI Bundle Reference Implementations`
+----------------------------------------
+
 .. autoclass:: MonaiAlgo
  :members:
+
+.. autoclass:: MonaiAlgoStats
+ :members:

docs/source/modules.md

@@ -270,15 +270,21 @@ A step-by-step [get started](https://github.com/Project-MONAI/tutorials/blob/mas
 
 ![federated-learning](../images/federated.svg)
 
-Using the MONAI bundle configurations, we can use MONAI's [`MonaiAlgo`](https://docs.monai.io/en/latest/fl.html#monai.fl.client.MonaiAlgo) class (an implementation of the abstract [`ClientAlgo`](https://docs.monai.io/en/latest/fl.html#clientalgo) class for federated learning)
+Using the MONAI bundle configurations, we can use MONAI's [`MonaiAlgo`](https://docs.monai.io/en/latest/fl.html#monai.fl.client.MonaiAlgo)
+class, an implementation of the abstract [`ClientAlgo`](https://docs.monai.io/en/latest/fl.html#clientalgo) class for federated learning (FL),
 to execute bundles from the [MONAI model zoo](https://github.com/Project-MONAI/model-zoo).
-Note that [`ClientAlgo`](https://docs.monai.io/en/latest/fl.html#clientalgo) is provided as an abstract base class for defining an algorithm to be run on any federated learning platform.
-[`MonaiAlgo`](https://docs.monai.io/en/latest/fl.html#monai.fl.client.MonaiAlgo) implements the main functionalities needed to run federated learning experiments, namely `train()`, `get_weights()`, and `evaluate()`.
+Note that [`ClientAlgo`](https://docs.monai.io/en/latest/fl.html#clientalgo) is provided as an abstract base class for
+defining an algorithm to be run on any federated learning platform.
+[`MonaiAlgo`](https://docs.monai.io/en/latest/fl.html#monai.fl.client.MonaiAlgo) implements the main functionalities needed
+to run federated learning experiments, namely `train()`, `get_weights()`, and `evaluate()`, that can be run using single- or multi-GPU training.
 On top, it provides implementations for life-cycle management of the component such as `initialize()`, `abort()`, and `finalize()`.
-
+The MONAI FL client also allows computing summary data statistics (e.g., intensity histograms) on the datasets defined in the bundle configs
+using the [`MonaiAlgoStats`](https://docs.monai.io/en/latest/fl.html#monai.fl.client.MonaiAlgoStats) class.
+These statistics can be shared and visualized on the FL server.
 [NVIDIA FLARE](https://github.com/NVIDIA/NVFlare), the federated learning platform developed by NVIDIA, has already built [the integration piece](https://github.com/NVIDIA/NVFlare/tree/2.2/integration/monai)
 with [`ClientAlgo`](https://docs.monai.io/en/latest/fl.html#clientalgo) to allow easy experimentation with MONAI bundles within their federated environment.
-[[Federated learning tutorial]](https://github.com/Project-MONAI/tutorials/tree/main/federated_learning/nvflare)
+Our [[federated learning tutorials]](https://github.com/Project-MONAI/tutorials/tree/main/federated_learning/nvflare) shows
+examples of single- & multi-GPU training and federated statistics workflows.
 
 ## Auto3dseg
 

docs/source/whatsnew_1_0.md

@@ -31,11 +31,16 @@ It has been tested on large-scale 3D medical imaging datasets in different modal
 ## Federated Learning Client
 ![federated-learning](../images/federated.svg)
 
-MONAI now includes the federated learning client algorithm APIs that are exposed as an abstract base class
+MONAI now includes the federated learning (FL) client algorithm APIs that are exposed as an abstract base class
 for defining an algorithm to be run on any federated learning platform.
 [NVIDIA FLARE](https://github.com/NVIDIA/NVFlare), the federated learning platform developed by [NVIDIA](https://www.nvidia.com/en-us/),
-has already built [the integration piece](https://github.com/NVIDIA/NVFlare/tree/2.2/integration/monai) with these new APIs.
-With [the new federated learning APIs](https://docs.monai.io/en/latest/fl.html), MONAI bundles can seamlessly be extended to a federated paradigm.
+has already built [the integration piece](https://github.com/NVIDIA/NVFlare/tree/dev/integration/monai) with these new APIs.
+With [the new federated learning APIs](https://docs.monai.io/en/latest/fl.html), MONAI bundles can seamlessly be extended to a federated paradigm
+and executed using single- or multi-GPU training.
+The MONAI FL client also allows computing summary data statistics (e.g., intensity histograms) on the datasets defined in the bundle configs.
+These can be shared and visualized on the FL server, for example, using NVIDIA FLARE's federated statistics operators,
+see [here](https://github.com/NVIDIA/NVFlare/tree/dev/integration/monai/examples/spleen_ct_segmentation) for an example.
+
 We welcome other federated learning toolkits to integrate with MONAI FL APIs, building a common foundation for
 collaborative learning in medical imaging.
 

monai/fl/client/client_algo.py

@@ -19,11 +19,11 @@ class BaseClient:
  Provide an abstract base class to allow the client to return summary statistics of the data.
 
  To define a new stats script, subclass this class and implement the
- following abstract methods:
+ following abstract methods::
 
  - self.get_data_stats()
 
- initialize(), abort(), and finalize() - inherited from `ClientAlgoStats` - can be optionally be implemented
+ initialize(), abort(), and finalize() -- inherited from `ClientAlgoStats`; can be optionally be implemented
  to help with lifecycle management of the class object.
  """
 
@@ -66,20 +66,23 @@ def get_data_stats(self, extra: Optional[dict] = None) -> ExchangeObject:
  For example, requested statistics.
 
  Returns:
+
  ExchangeObject: summary statistics.
 
- Extra dict example:
+ Extra dict example::
+
  requested_stats = {
  FlStatistics.STATISTICS: metrics,
  FlStatistics.NUM_OF_BINS: num_of_bins,
  FlStatistics.BIN_RANGES: bin_ranges
  }
 
- Returned ExchangeObject example:
+ Returned ExchangeObject example::
 
  ExchangeObject(
  statistics = {...}
  )
+
  """
  raise NotImplementedError(f"Subclass {self.__class__.__name__} must implement this method.")