[Documentation] Add SM Training Compiler node and clean up TOC (aws#3041)

* Add a note to remove account ids when reporting bugs with logs

* add sm training compiler node, delete duplicated pages

* Add topics

* clean up titles and TOC

Co-authored-by: Miyoung Choi <myoung8739@gamil.com>

Author: mchoi8739

index.rst

@@ -93,16 +93,25 @@ On SageMaker Studio, you will need to open a terminal, go to your home folder, t
    sagemaker-featurestore/index
 
 
+.. toctree::
+   :maxdepth: 1
+   :caption: Frameworks
+
+   training/frameworks
+
+
 .. toctree::
    :maxdepth: 1
    :caption: Training
 
-   training/index
    training/algorithms
-   training/frameworks
+   reinforcement_learning/index
+   sagemaker-experiments/index
    sagemaker-debugger/index
    training/tuning
    training/distributed_training/index
+   sagemaker-training-compiler/index
+   sagemaker-script-mode/index
    training/bring_your_own_container
    training/management
 
@@ -114,13 +123,6 @@ On SageMaker Studio, you will need to open a terminal, go to your home folder, t
    inference/index
 
 
-.. toctree::
-   :maxdepth: 1
-   :caption: Frameworks
-
-   training/frameworks
-
-
 .. toctree::
    :maxdepth: 1
    :caption: Workflows

reinforcement_learning/index.rst

@@ -1,3 +1,6 @@
+######################
+Reinforcement Learning
+######################
 
 .. image:: rl_mountain_car_coach_gymEnv/successful_policy.gif
   :width: 600

sagemaker-experiments/index.rst

@@ -1,3 +1,8 @@
+###########
+Experiments
+###########
+
+
 Get started with Experiments
 ============================
 

sagemaker-script-mode/index.rst

@@ -1,3 +1,7 @@
+###########
+Script Mode
+###########
+
 SageMaker Script Mode Examples
 ==============================
 
@@ -11,4 +15,4 @@ SageMaker Script Mode at Increasing Levels of Customization
 .. toctree::
    :maxdepth: 1
 
-   sagemaker-script-mode
+   sagemaker-script-mode

sagemaker-training-compiler/index.rst

@@ -0,0 +1,41 @@
+#################
+Training Compiler
+#################
+
+Train deep learning (DL) models faster on scalable GPU instances managed by SageMaker.
+
+
+Get started with SageMaker Training Compiler
+============================================
+
+PyTorch with Hugging Face Transformers
+--------------------------------------
+
+For single-node single-GPU training:
+
+.. toctree::
+   :maxdepth: 1
+
+   huggingface/pytorch_single_gpu_single_node/albert-base-v2/albert-base-v2
+   huggingface/pytorch_single_gpu_single_node/bert-base-cased/bert-base-cased-single-node-single-gpu
+   huggingface/pytorch_single_gpu_single_node/roberta-base/roberta-base
+
+
+For single-node multi-GPU training:
+
+.. toctree::
+   :maxdepth: 1
+
+   huggingface/pytorch_multiple_gpu_single_node/language-modeling-multi-gpu-single-node
+
+
+For multi-node multi-GPU training:
+
+.. toctree::
+   :maxdepth: 1
+
+   huggingface/pytorch_multiple_gpu_multiple_node/language-modeling-multi-gpu-multi-node
+
+
+For more information, see `Amazon SageMaker Training Compiler <https://docs.aws.amazon.com/sagemaker/latest/dg/training-compiler.html>`__
+in the Amazon SageMaker Developer Guide.