[tune] horovod trainable (#10304)

This commit is contained in:
Richard Liaw
2020-09-03 16:53:35 -07:00
committed by GitHub
parent 7068c63dd8
commit 43a7a64b30
11 changed files with 485 additions and 2 deletions
+2
View File
@@ -36,6 +36,8 @@ MOCK_MODULES = [
"blist",
"gym",
"gym.spaces",
"horovod",
"horovod.ray",
"kubernetes",
"psutil",
"ray._raylet",
+8
View File
@@ -43,6 +43,14 @@ Torch (tune.integration.torch)
.. autofunction:: ray.tune.integration.torch.is_distributed_trainable
.. _tune-integration-horovod:
Horovod (tune.integration.horovod)
----------------------------------
.. autofunction:: ray.tune.integration.horovod.DistributedTrainableCreator
.. _tune-integration-wandb:
Weights and Biases (tune.integration.wandb)
@@ -0,0 +1,6 @@
:orphan:
horovod_simple
~~~~~~~~~~~~~~
.. literalinclude:: /../../python/ray/tune/examples/horovod_simple.py
+4
View File
@@ -35,6 +35,10 @@ Tensorflow/Keras Examples
- :doc:`/tune/examples/pbt_memnn_example`: Example of training a Memory NN on bAbI with Keras using PBT.
- :doc:`/tune/examples/tf_mnist_example`: Converts the Advanced TF2.0 MNIST example to use Tune with the Trainable. This uses `tf.function`. Original code from tensorflow: https://www.tensorflow.org/tutorials/quickstart/advanced
Horovod Example
~~~~~~~~~~~~~~~
- :doc:`/tune/examples/horovod_simple`: Leverages the :ref:`Horovod-Tune <tune-integration-horovod>` integration to launch a distributed training + tuning job.
PyTorch Examples
~~~~~~~~~~~~~~~~