[docs] [rllib] Remove outdated RLlib examples from the docs 1/N

doc/source/ray-overview/examples.rst

@@ -639,66 +639,12 @@ Ray Examples
 
         Example of how to setup an RLlib algorithm against a locally running Unity3D editor
 
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item rl
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/env_rendering_and_recording.py
-
-        Rendering and recording of an environment
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item rl
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/coin_game_env.py
-
-        Coin game example with RLlib
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item reinforcement-learning rllib
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/dmlab_watermaze.py
-
-        Example for how to use a DMLab environment (Watermaze)
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item rl
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/recommender_system_with_recsim_and_slateq.py
-
-        RecSym environment example (for recommender systems) using the SlateQ algorithm
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item reinforcement-learning rllib
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/sumo_env_local.py
-
-        Example demonstrating how to use the SUMO simulator in connection with RLlib.
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item rl
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/vizdoom_with_attention_net.py
-
-        VizDoom example script using RLlib's auto-attention wrapper
-
     .. grid-item-card:: :bdg-secondary:`Code example`
         :class-item: gallery-item rl
         :link: https://github.com/ray-project/ray/blob/master/rllib/env/tests/test_env_with_subprocess.py
 
         Example of how to ensure subprocesses spawned by envs are killed when RLlib exits.
 
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item rl
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/attention_net.py
-
-        Attention Net (GTrXL) learning the "repeat-after-me" environment
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item reinforcement-learning rllib
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/lstm_auto_wrapping.py
-
-        Example showing how to use the auto-LSTM wrapper for your default- and custom models in RLlib.
-
-    .. grid-item-card:: :bdg-secondary:`Code example`
-        :class-item: gallery-item reinforcement-learning rllib pytorch tensorflow
-        :link: https://github.com/ray-project/ray/blob/master/rllib/examples/custom_rnn_model.py
-
-        Example of using a custom Keras- or PyTorch RNN model.
-
     .. grid-item-card:: :bdg-secondary:`Code example`
         :class-item: gallery-item reinforcement-learning rllib
         :link: https://github.com/ray-project/ray/blob/master/rllib/examples/custom_model_loss_and_metrics.py

doc/source/rllib/feature_overview.rst

@@ -35,25 +35,6 @@
         on 100s of CPUs/nodes thus parallelizing and speeding up learning.
 
 
-.. container:: clear-both
-
-    .. container:: buttons-float-left
-
-        .. https://docs.google.com/drawings/d/1b8uaRo0KjPH-x-elBmyvDwAA4I2oy8cj3dxNnUT3HTE/edit
-
-        .. image:: images/sigils/rllib-sigil-vector-envs.svg
-            :width: 100
-            :target: https://github.com/ray-project/ray/blob/master/rllib/examples/env_rendering_and_recording.py
-
-    .. container::
-
-        **Vectorized (batched) and remote (parallel) environments**: RLlib auto-vectorizes
-        your ``gym.Envs`` via the ``num_envs_per_worker`` config. Environment workers can
-        then batch and thus significantly speedup the action computing forward pass.
-        On top of that, RLlib offers the ``remote_worker_envs`` config to create
-        `single environments (within a vectorized one) as ray Actors <https://github.com/ray-project/ray/blob/master/rllib/examples/remote_envs_with_inference_done_on_main_node.py>`_,
-        thus parallelizing even the env stepping process.
-
 .. container:: clear-both
 
     .. container:: buttons-float-left

doc/source/rllib/rllib-examples.rst

@@ -43,33 +43,15 @@ Environments and Adapters
    Use this example to try things out and watch the game and the learning progress live in the editor.
    Providing a compiled game, this example could also run in distributed fashion with `num_workers > 0`.
    For a more heavy-weight, distributed, cloud-based example, see ``Unity3D client/server`` below.
-- `Rendering and recording of an environment <https://github.com/ray-project/ray/blob/master/rllib/examples/env_rendering_and_recording.py>`__:
-   Example showing how to switch on rendering and recording of an env.
-- `Coin Game Example <https://github.com/ray-project/ray/blob/master/rllib/examples/coin_game_env.py>`__:
-   Coin Game Env Example (provided by the "Center on Long Term Risk").
-- `DMLab Watermaze example <https://github.com/ray-project/ray/blob/master/rllib/examples/dmlab_watermaze.py>`__:
-   Example for how to use a DMLab environment (Watermaze).
-- `RecSym environment example (for recommender systems) using the SlateQ algorithm <https://github.com/ray-project/ray/blob/master/rllib/examples/recommender_system_with_recsim_and_slateq.py>`__:
-   Script showing how to train SlateQ on a RecSym environment.
-- `SUMO (Simulation of Urban MObility) environment example <https://github.com/ray-project/ray/blob/master/rllib/examples/sumo_env_local.py>`__:
-   Example demonstrating how to use the SUMO simulator in connection with RLlib.
-- `VizDoom example script using RLlib's auto-attention wrapper <https://github.com/ray-project/ray/blob/master/rllib/examples/vizdoom_with_attention_net.py>`__:
-   Script showing how to run PPO with an attention net against a VizDoom gym environment.
 - `Subprocess environment <https://github.com/ray-project/ray/blob/master/rllib/env/tests/test_env_with_subprocess.py>`__:
    Example of how to ensure subprocesses spawned by envs are killed when RLlib exits.
 
 
 Custom- and Complex Models
 --------------------------
 
-- `Attention Net (GTrXL) learning the "repeat-after-me" environment <https://github.com/ray-project/ray/blob/master/rllib/examples/attention_net.py>`__:
-   Example showing how to use the auto-attention wrapper for your default- and custom models in RLlib.
-- `LSTM model learning the "repeat-after-me" environment <https://github.com/ray-project/ray/blob/master/rllib/examples/lstm_auto_wrapping.py>`__:
-   Example showing how to use the auto-LSTM wrapper for your default- and custom models in RLlib.
 - `Custom Keras model <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_keras_model.py>`__:
    Example of using a custom Keras model.
-- `Custom Keras/PyTorch RNN model <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_rnn_model.py>`__:
-   Example of using a custom Keras- or PyTorch RNN model.
 - `Registering a custom model with supervised loss <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_model_loss_and_metrics.py>`__:
    Example of defining and registering a custom model with a supervised loss.
 - `Batch normalization <https://github.com/ray-project/ray/blob/master/rllib/examples/batch_norm_model.py>`__:

doc/source/rllib/rllib-models.rst

@@ -300,20 +300,6 @@ More examples and explanations on how to implement custom Tuple/Dict processing
 custom RNNs, custom model APIs (on top of default models) follow further below.
 
 
-Wrapping a Custom Model (TF and PyTorch) with an LSTM- or Attention Net
-```````````````````````````````````````````````````````````````````````
-
-You can also use a custom (TF or PyTorch) model with our auto-wrappers for LSTMs (``use_lstm=True``) or Attention networks (``use_attention=True``).
-For example, if you would like to wrap some non-default model logic with an LSTM, simply do:
-
-.. literalinclude:: ../../../rllib/examples/lstm_auto_wrapping.py
-   :language: python
-   :start-after: __sphinx_doc_begin__
-   :end-before: __sphinx_doc_end__
-
-
-.. _rnns:
-
 Implementing custom Recurrent Networks
 ``````````````````````````````````````
 
@@ -351,8 +337,7 @@ Check out RLlib's `GTrXL (Attention Net) <https://arxiv.org/abs/1910.06764>`__ i
 to get a better idea on how to write your own models of this type. These are the models we use
 as wrappers when ``use_attention=True``.
 
-You can run `this example script <https://github.com/ray-project/ray/blob/master/rllib/examples/attention_net.py>`__ to run these nets within some of our algorithms.
-`There is also a test case <https://github.com/ray-project/ray/blob/master/rllib/models/tests/test_attention_nets.py>`__, which confirms their learning capabilities in PPO and IMPALA.
+This `test case <https://github.com/ray-project/ray/blob/master/rllib/models/tests/test_attention_nets.py>`__ confirms their learning capabilities in PPO and IMPALA.
 
 Batch Normalization
 ```````````````````

release/rllib_tests/unit_gpu_tests/run.sh

@@ -22,9 +22,6 @@ SUCCESS=1
 # Run all test cases, but with a forced num_gpus=1.
 export RLLIB_NUM_GPUS=1
 
-if python rllib/examples/attention_net.py --as-test --stop-reward=20 --num-cpus=0; then SUCCESS=0; fi
-if python rllib/examples/attention_net.py --framework=torch --as-test --stop-reward=20 --num-cpus=0; then SUCCESS=0; fi
-
 popd || true
 
 echo "{'passed': $SUCCESS}" > "${TEST_OUTPUT_JSON:-/tmp/release_test_out.json}"

rllib/BUILD

@@ -739,13 +739,6 @@ py_test(
 # Tag: env
 # --------------------------------------------------------------------
 
-py_test(
-    name = "env/tests/test_env_with_subprocess",
-    tags = ["team:rllib", "env"],
-    size = "medium",
-    srcs = ["env/tests/test_env_with_subprocess.py"]
-)
-
 sh_test(
     name = "env/tests/test_local_inference_cartpole",
     tags = ["team:rllib", "env"],
@@ -1907,24 +1900,6 @@ py_test(
     args = ["--stop-iter=2", "--framework=torch"]
 )
 
-py_test(
-    name = "examples/attention_net_tf",
-    main = "examples/attention_net.py",
-    tags = ["team:rllib", "exclusive", "examples"],
-    size = "medium",
-    srcs = ["examples/attention_net.py"],
-    args = ["--as-test", "--stop-reward=70", "--framework=tf"]
-)
-
-py_test(
-    name = "examples/attention_net_torch",
-    main = "examples/attention_net.py",
-    tags = ["team:rllib", "exclusive", "examples"],
-    size = "medium",
-    srcs = ["examples/attention_net.py"],
-    args = ["--as-test", "--stop-reward=70", "--framework torch"]
-)
-
 py_test(
     name = "examples/autoregressive_action_dist_tf",
     main = "examples/autoregressive_action_dist.py",