Use master for links to docs in source (#10866)

Resolved Conflicts:
        .github/PULL_REQUEST_TEMPLATE.md
        rllib/agents/dqn/apex.py
        rllib/agents/dqn/dqn.py

.github/PULL_REQUEST_TEMPLATE.md

@@ -13,7 +13,7 @@
 ## Checks
 
 - [ ] I've run `scripts/format.sh` to lint the changes in this PR.
-- [ ] I've included any doc changes needed for https://docs.ray.io/en/latest/.
+- [ ] I've included any doc changes needed for https://docs.ray.io/en/master/.
 - [ ] I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failure rates at https://ray-travis-tracker.herokuapp.com/.
 - Testing Strategy
    - [ ] Unit tests

README.rst

@@ -16,7 +16,7 @@ Ray is packaged with the following libraries for accelerating machine learning w
 - `Ray Serve`_: Scalable and Programmable Serving
 
 Install Ray with: ``pip install ray``. For nightly wheels, see the
-`Installation page <https://docs.ray.io/en/latest/installation.html>`__.
+`Installation page <https://docs.ray.io/en/master/installation.html>`__.
 
 **NOTE:** As of Ray 0.8.1, Python 2 is no longer supported.
 

python/ray/setup-dev.py

@@ -63,7 +63,7 @@ if __name__ == "__main__":
     print("Created links.\n\nIf you run into issues initializing Ray, please "
           "ensure that your local repo and the installed Ray are in sync "
           "(pip install -U the latest wheels at "
-          "https://docs.ray.io/en/latest/installation.html, "
+          "https://docs.ray.io/en/master/installation.html, "
           "and ensure you are up-to-date on the master branch on git).\n\n"
           "Note that you may need to delete the package symlinks when pip "
           "installing new Ray versions to prevent pip from overwriting files "

python/ray/tests/test_metrics.py

@@ -402,7 +402,7 @@ def test_memory_dashboard(shutdown_only):
     """Test Memory table.
 
     These tests verify examples in this document.
-    https://docs.ray.io/en/latest/memory-management.html#debugging-using-ray-memory
+    https://docs.ray.io/en/master/memory-management.html#debugging-using-ray-memory
     """
     addresses = ray.init(num_cpus=2)
     webui_url = addresses["webui_url"].replace("127.0.0.1", "http://127.0.0.1")

rllib/README.md

@@ -27,4 +27,4 @@ If you've found RLlib useful for your research, you can cite the [paper](https:/
 Development Install
 -------------------
 
-You can develop RLlib locally without needing to compile Ray by using the [setup-dev.py](https://github.com/ray-project/ray/blob/master/python/ray/setup-dev.py) script. This sets up links between the ``rllib`` dir in your git repo and the one bundled with the ``ray`` package. When using this script, make sure that your git branch is in sync with the installed Ray binaries (i.e., you are up-to-date on [master](https://github.com/ray-project/ray) and have the latest [wheel](https://docs.ray.io/en/latest/installation.html) installed.)
+You can develop RLlib locally without needing to compile Ray by using the [setup-dev.py](https://github.com/ray-project/ray/blob/master/python/ray/setup-dev.py) script. This sets up links between the ``rllib`` dir in your git repo and the one bundled with the ``ray`` package. When using this script, make sure that your git branch is in sync with the installed Ray binaries (i.e., you are up-to-date on [master](https://github.com/ray-project/ray) and have the latest [wheel](https://docs.ray.io/en/master/installation.html) installed.)

rllib/agents/dqn/apex.py

@@ -1,3 +1,17 @@
+"""
+Distributed Prioritized Experience Replay (Ape-X)
+=================================================
+
+This file defines a DQN trainer using the Ape-X architecture.
+
+Ape-X uses a single GPU learner and many CPU workers for experience collection.
+Experience collection can scale to hundreds of CPU workers due to the
+distributed prioritization of experience prior to storage in replay buffers.
+
+Detailed documentation:
+https://docs.ray.io/en/master/rllib-algorithms.html#distributed-prioritized-experience-replay-ape-x
+"""  # noqa: E501
+
 import collections
 import copy
 

rllib/agents/dqn/dqn.py

@@ -1,3 +1,14 @@
+"""
+Deep Q-Networks (DQN, Rainbow, Parametric DQN)
+==============================================
+
+This file defines the distributed Trainer class for the Deep Q-Networks
+algorithm. See `dqn_[tf|torch]_policy.py` for the definition of the policies.
+
+Detailed documentation:
+https://docs.ray.io/en/master/rllib-algorithms.html#deep-q-networks-dqn-rainbow-parametric-dqn
+"""  # noqa: E501
+
 import logging
 from typing import Type
 

rllib/agents/pg/pg.py

@@ -5,7 +5,7 @@ Policy Gradient (PG)
 This file defines the distributed Trainer class for policy gradients.
 See `pg_[tf|torch]_policy.py` for the definition of the policy loss.
 
-Detailed documentation: https://docs.ray.io/en/latest/rllib-algorithms.html#pg
+Detailed documentation: https://docs.ray.io/en/master/rllib-algorithms.html#pg
 """
 
 from typing import Optional, Type

rllib/agents/ppo/README.md

@@ -5,19 +5,19 @@ Implementations of:
 
 1) Proximal Policy Optimization (PPO). 
 
-    **[Detailed Documentation](https://docs.ray.io/en/latest/rllib-algorithms.html#ppo)**
+    **[Detailed Documentation](https://docs.ray.io/en/master/rllib-algorithms.html#ppo)**
 
     **[Implementation](https://github.com/ray-project/ray/blob/master/rllib/agents/ppo/ppo.py)**
 
 2) Asynchronous Proximal Policy Optimization (APPO).
 
-    **[Detailed Documentation](https://docs.ray.io/en/latest/rllib-algorithms.html#appo)**
+    **[Detailed Documentation](https://docs.ray.io/en/master/rllib-algorithms.html#appo)**
 
     **[Implementation](https://github.com/ray-project/ray/blob/master/rllib/agents/ppo/appo.py)**
 
 3) Decentralized Distributed Proximal Policy Optimization (DDPPO)
 
-    **[Detailed Documentation](https://docs.ray.io/en/latest/rllib-algorithms.html#decentralized-distributed-proximal-policy-optimization-dd-ppo)**
+    **[Detailed Documentation](https://docs.ray.io/en/master/rllib-algorithms.html#decentralized-distributed-proximal-policy-optimization-dd-ppo)**
 
     **[Implementation](https://github.com/ray-project/ray/blob/master/rllib/agents/ppo/ddppo.py)**
 

rllib/agents/ppo/appo.py

@@ -7,7 +7,7 @@ of proximal policy optimization (APPO).
 See `appo_[tf|torch]_policy.py` for the definition of the policy loss.
 
 Detailed documentation:
-https://docs.ray.io/en/latest/rllib-algorithms.html#appo
+https://docs.ray.io/en/master/rllib-algorithms.html#appo
 """
 from typing import Optional, Type
 

rllib/agents/ppo/ppo.py

@@ -6,7 +6,7 @@ This file defines the distributed Trainer class for proximal policy
 optimization.
 See `ppo_[tf|torch]_policy.py` for the definition of the policy loss.
 
-Detailed documentation: https://docs.ray.io/en/latest/rllib-algorithms.html#ppo
+Detailed documentation: https://docs.ray.io/en/master/rllib-algorithms.html#ppo
 """
 
 import logging