[docs] Render ActorPool documentation, etc (#9433)

This commit is contained in:
Richard Liaw 2020-07-13 17:59:22 -07:00 committed by GitHub
parent 6ad13e0da8
commit 7abf7a0109
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 127 additions and 31 deletions

View file

@ -1,8 +1,7 @@
Ray Tutorials and Examples
==========================
Get started with Ray, Tune, and RLlib with these notebooks that you can run online in CoLab or Binder:
* `Ray Tutorial Notebooks <https://github.com/ray-project/tutorial>`__
Get started with Ray, Tune, and RLlib with these notebooks that you can run online in CoLab or Binder: `Ray Tutorial Notebooks <https://github.com/ray-project/tutorial>`__
.. toctree::
:hidden:
@ -17,9 +16,6 @@ Get started with Ray, Tune, and RLlib with these notebooks that you can run onli
plot_example-lm.rst
Example Gallery
---------------
.. customgalleryitem::
:tooltip: Build a simple parameter server using Ray.
:figure: /images/param_actor.png

View file

@ -1,51 +1,168 @@
Ray Package Reference
=====================
Python API
----------
.. _ray-init-ref:
ray.init
~~~~~~~~
.. autofunction:: ray.init
.. _ray-is_initialized-ref:
ray.is_initialized
~~~~~~~~~~~~~~~~~~
.. autofunction:: ray.is_initialized
.. _ray-remote-ref:
ray.remote
~~~~~~~~~~
.. autofunction:: ray.remote
.. _ray-get-ref:
ray.get
~~~~~~~
.. autofunction:: ray.get
.. _ray-wait-ref:
ray.wait
~~~~~~~~
.. autofunction:: ray.wait
.. _ray-put-ref:
ray.put
~~~~~~~
.. autofunction:: ray.put
.. _ray-kill-ref:
ray.kill
~~~~~~~~
.. autofunction:: ray.kill
.. _ray-cancel-ref:
ray.cancel
~~~~~~~~~~
.. autofunction:: ray.cancel
.. _ray-get_gpu_ids-ref:
ray.get_gpu_ids
~~~~~~~~~~~~~~~
.. autofunction:: ray.get_gpu_ids
.. _ray-get_resource_ids-ref:
ray.get_resource_ids
~~~~~~~~~~~~~~~~~~~~
.. autofunction:: ray.get_resource_ids
.. _ray-get_webui_url-ref:
ray.get_webui_url
~~~~~~~~~~~~~~~~~
.. autofunction:: ray.get_webui_url
.. _ray-shutdown-ref:
ray.shutdown
~~~~~~~~~~~~
.. autofunction:: ray.shutdown
.. _ray-register_custom_serializer-ref:
ray.register_custom_serializer
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: ray.register_custom_serializer
.. _ray-profile-ref:
ray.profile
~~~~~~~~~~~
.. autofunction:: ray.profile
.. _ray-method-ref:
ray.method
~~~~~~~~~~
.. autofunction:: ray.method
Inspect the Cluster State
-------------------------
ray.util.ActorPool
~~~~~~~~~~~~~~~~~~
.. autoclass:: ray.util.ActorPool
:members:
.. _ray-nodes-ref:
ray.nodes
~~~~~~~~~
.. autofunction:: ray.nodes
.. _ray-objects-ref:
ray.objects
~~~~~~~~~~~
.. autofunction:: ray.objects
.. _ray-timeline-ref:
ray.timeline
~~~~~~~~~~~~
.. autofunction:: ray.timeline
.. _ray-object_transfer_timeline-ref:
ray.object_transfer_timeline
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: ray.object_transfer_timeline
.. _ray-cluster_resources-ref:
ray.cluster_resources
~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: ray.cluster_resources
.. _ray-available_resources-ref:
ray.available_resources
~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: ray.available_resources
.. _ray-errors-ref:
ray.errors
~~~~~~~~~~
.. autofunction:: ray.errors
Experimental APIs

View file

@ -85,10 +85,10 @@ Tune Quick Start
.. code-block:: bash
$ pip install ray torch torchvision filelock
$ pip install 'ray[tune]'
This example runs a small grid search to train a CNN using PyTorch and Tune.
This example runs a small grid search with an iterative training function.
.. literalinclude:: ../../../python/ray/tune/tests/example.py
:language: python

View file

@ -65,12 +65,7 @@ Serialization notes
Last resort: Custom Serialization
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
If none of these options work, you can try registering a custom serializer.
.. autofunction:: ray.register_custom_serializer
:noindex:
Below is an example of using ``ray.register_custom_serializer``:
If none of these options work, you can try registering a custom serializer with ``ray.register_custom_serializer`` (:ref:`docstring <ray-register_custom_serializer-ref>`):
.. code-block:: python

View file

@ -172,7 +172,7 @@ Further, remote functions can return multiple object refs.
a_id, b_id, c_id = return_multiple.remote()
Remote functions can be canceled by calling ``ray.cancel`` on the returned Object ref. Remote actor functions can be stopped by killing the actor using the ``ray.kill`` interface.
Remote functions can be canceled by calling ``ray.cancel`` (:ref:`docstring <ray-cancel-ref>`) on the returned Object ref. Remote actor functions can be stopped by killing the actor using the ``ray.kill`` interface.
.. code-block:: python
@ -184,9 +184,6 @@ Remote functions can be canceled by calling ``ray.cancel`` on the returned Objec
obj_ref = blocking_operation.remote()
ray.cancel(obj_ref)
.. autofunction:: ray.cancel
:noindex:
Objects in Ray
--------------
@ -199,16 +196,13 @@ similar.
Object refs can be created in multiple ways.
1. They are returned by remote function calls.
2. They are returned by ``ray.put``.
2. They are returned by ``ray.put`` (:ref:`docstring <ray-put-ref>`).
.. code-block:: python
y = 1
object_ref = ray.put(y)
.. autofunction:: ray.put
:noindex:
.. note::
Remote objects are immutable. That is, their values cannot be changed after
@ -219,7 +213,7 @@ Object refs can be created in multiple ways.
Fetching Results
----------------
The command ``ray.get(x_id, timeout=None)`` takes an object ref and creates a Python object
The command ``ray.get(x_id, timeout=None)`` (:ref:`docstring <ray-get-ref>`) takes an object ref and creates a Python object
from the corresponding remote object. First, if the current node's object store
does not contain the object, the object is downloaded. Then, if the object is a `numpy array <https://docs.scipy.org/doc/numpy/reference/generated/numpy.array.html>`__
or a collection of numpy arrays, the ``get`` call is zero-copy and returns arrays backed by shared object store memory.
@ -247,21 +241,15 @@ You can also set a timeout to return early from a ``get`` that's blocking for to
except RayTimeoutError:
print("`get` timed out.")
.. autofunction:: ray.get
:noindex:
After launching a number of tasks, you may want to know which ones have
finished executing. This can be done with ``ray.wait``. The function
finished executing. This can be done with ``ray.wait`` (:ref:`ray-wait-ref`). The function
works as follows.
.. code:: python
ready_ids, remaining_ids = ray.wait(object_refs, num_returns=1, timeout=None)
.. autofunction:: ray.wait
:noindex:
Object Eviction
---------------