wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-08-12 12:20:11 +08:00
Code Issues Packages Projects Releases Wiki Activity

[tune] cleanup error messaging/diagnose_serialization helper (#10210)

Browse Source
This commit is contained in:
Richard Liaw
2020-08-22 11:50:49 -07:00
committed by GitHub
parent 24ee496b89
commit 6bd5458bef
8 changed files with 178 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/source/tune/user-guide.rst
+2 -2
View File
@@ -197,7 +197,7 @@ Distributed Checkpointing
On a multinode cluster, Tune automatically creates a copy of all trial checkpoints on the head node. This requires the Ray cluster to be started with the :ref:`cluster launcher <ref-automatic-cluster>` and also requires rsync to be installed.
Note that you must use the ``tune.checkpoint_dir`` API to trigger syncing. Also, if running Tune on Kubernetes, be sure to use the :ref:`KubernetesSyncer <tune-kubernetes>` to transfer files between different pods.
Note that you must use the ``tune.checkpoint_dir`` API to trigger syncing. Also, if running Tune on Kubernetes, be sure to use the :ref:`KubernetesSyncer <tune-kubernetes>` to transfer files between different pods.
If you do not use the cluster launcher, you should set up a NFS or global file system and
disable cross-node syncing:
@@ -225,7 +225,7 @@ You often will want to compute a large object (e.g., training data, model weight
# X_id can be referenced in closures
X_id = pin_in_object_store(np.random.random(size=100000000))
def f(config, reporter):
def f(config):
X = get_pinned_object(X_id)
# use X