File tree 1 file changed +6
-6
lines changed
google/cloud/aiplatform/preview/vertex_ray
1 file changed +6
-6
lines changed Original file line number Diff line number Diff line change @@ -70,17 +70,17 @@ def create_ray_cluster(
70
70
)]
71
71
72
72
cluster_resource_name = vertex_ray.create_ray_cluster(
73
- head_node_type=head_node_type,
74
- network="my-vpc",
75
- worker_node_types=worker_node_types,
73
+ head_node_type=head_node_type,
74
+ network="projects/ my-project-number/global/networks/my- vpc-name ",
75
+ worker_node_types=worker_node_types,
76
76
)
77
77
78
78
After a ray cluster is set up, you can call
79
- `ray.init(vertex_ray://{cluster_resource_name}, runtime_env=...)` without
79
+ `ray.init(f" vertex_ray://{cluster_resource_name}" , runtime_env=...)` without
80
80
specifying ray cluster address to connect to the cluster. To shut down the
81
81
cluster you can call `ray.delete_ray_cluster()`.
82
- Note: If the active ray cluster haven't shut down, you cannot create a new ray
83
- cluster with the same cluster_name.
82
+ Note: If the active ray cluster has not finished shutting down, you cannot
83
+ create a new ray cluster with the same cluster_name.
84
84
85
85
Args:
86
86
head_node_type: The head node resource. Resources.node_count must be 1.
You can’t perform that action at this time.
0 commit comments