documentation: the SageMaker distributed data parallel v1.4.0 release

.readthedocs.yml

@@ -5,7 +5,7 @@
 version: 2
 
 python:
-  version: 3.6
+  version: 3.9
   install:
     - method: pip
       path: .

doc/api/training/distributed.rst

@@ -4,8 +4,25 @@ SageMaker distributed training libraries offer both data parallel and model para
 They combine software and hardware technologies to improve inter-GPU and inter-node communications.
 They extend SageMaker’s training capabilities with built-in options that require only small code changes to your training scripts.
 
+.. _sdp_api_docs_toc:
+
+The SageMaker Distributed Data Parallel Library
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. toctree::
+    :maxdepth: 3
+
+    smd_data_parallel
+    sdp_versions/latest
+    smd_data_parallel_use_sm_pysdk
+    smd_data_parallel_release_notes/smd_data_parallel_change_log
+
+.. _smp_api_docs_toc:
+
+The SageMaker Distributed Model Parallel Library
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. toctree::
    :maxdepth: 3
 
-   smd_data_parallel
    smd_model_parallel

doc/api/training/sdp_versions/archives.rst

@@ -0,0 +1,8 @@
+.. _smddp-version-archive:
+
+.. toctree::
+    :maxdepth: 1
+
+    v1_2_x.rst
+    v1_1_x.rst
+    v1_0_0.rst

doc/api/training/sdp_versions/latest.rst

@@ -1,9 +1,47 @@
+.. _sdp_api_docs:
 
-Version 1.2.x (Latest)
+#############################################
+Use the Library to Adapt Your Training Script
+#############################################
+
+This section contains the SageMaker distributed data parallel API documentation.
+If you are a new user of this library, it is recommended you use this guide alongside
+`SageMaker's Distributed Data Parallel Library
+<https://docs.aws.amazon.com/sagemaker/latest/dg/data-parallel.html>`_.
+
+The library provides framework-specific APIs for TensorFlow and PyTorch.
+
+Select the latest or one of the previous versions of the API documentation
+depending on the version of the library you use.
+
+.. important::
+
+   The distributed data parallel library supports training jobs using CUDA 11 or later.
+   When you define a :class:`sagemaker.tensorflow.estimator.TensorFlow` or
+   :class:`sagemaker.pytorch.estimator.PyTorch`
+   estimator with the data parallel library enabled,
+   SageMaker uses CUDA 11. When you extend or customize your own training image,
+   you must use a base image with CUDA 11 or later. See
+   `SageMaker Python SDK's distributed data parallel library APIs
+   <https://docs.aws.amazon.com/sagemaker/latest/dg/data-parallel-use-api.html#data-parallel-use-python-skd-api>`_
+   for more information.
+
+Version 1.4.0 (Latest)
 ======================
 
 .. toctree::
    :maxdepth: 1
 
-   latest/smd_data_parallel_pytorch.rst
-   latest/smd_data_parallel_tensorflow.rst
+   latest/smd_data_parallel_pytorch
+   latest/smd_data_parallel_tensorflow
+
+Documentation Archive
+=====================
+
+To find the API documentation for the previous versions of the library,
+choose one of the following:
+
+.. toctree::
+   :maxdepth: 1
+
+   archives