You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
You can spin up the [ArangoML](https://github.com/arangoml) engine on existing ArangoDeployment.
@@ -29,7 +29,7 @@ If you don't have one yet, consider checking [kube-arangodb installation guide](
29
29
- Create Kubernetes Secret containing CA certificates to validate connection to endpoint if your Minio installation uses encrypted connection. The secret data should contain two fields: `ca.crt` and `ca.key` (both PEM-encoded).
30
30
- Create ArangoMLStorage resource. Example:
31
31
```yaml
32
-
apiVersion: ml.arangodb.com/v1alpha1
32
+
apiVersion: ml.arangodb.com/v1beta1
33
33
kind: ArangoMLStorage
34
34
metadata:
35
35
name: myarangoml-storage
@@ -50,7 +50,7 @@ If you don't have one yet, consider checking [kube-arangodb installation guide](
50
50
3) Create `ArangoMLExtension` CR. The name of extension **must** be the same as the name of `ArangoDeployment` and it should be created in the same namespace.
51
51
Assuming you have ArangoDeployment with name `myarangodb`, create CR:
52
52
```yaml
53
-
apiVersion: ml.arangodb.com/v1alpha1
53
+
apiVersion: ml.arangodb.com/v1beta1
54
54
kind: ArangoMLExtension
55
55
metadata:
56
56
name: myarangodb
@@ -59,16 +59,9 @@ If you don't have one yet, consider checking [kube-arangodb installation guide](
59
59
name: myarangoml-storage # name of the ArangoMLStorage created on the previous step
60
60
deployment:
61
61
# you can add here: tolerations, nodeSelector, nodeAffinity, scheduler and many other parameters. See full CRD reference for details.
62
-
replicas: 1 # by default only one pod is running which contains containers for each component (prediction, training, project). You can scale it up or down.
63
-
prediction:
64
-
image: <prediction-image>
65
-
# you can configure various parameters for container running this component here. See full CRD reference for details.
66
-
project:
67
-
image: <projects-image>
68
-
# you can configure various parameters for container running this component here. See full CRD reference for details.
69
-
training:
70
-
image: <training-image>
71
-
# you can configure various parameters for container running this component here. See full CRD reference for details.
62
+
replicas: 1 # by default only one pod is running which contains containers for api. You can scale it up or down.
63
+
image: <api-image>
64
+
# you can configure various parameters for container running this component here. See full CRD reference for details.
72
65
init: # configuration for Kubernetes Job running initial bootstrap of ArangoML for your cluster.
73
66
image: <init-image>
74
67
# you can add here: tolerations, nodeSelector, nodeAffinity, scheduler and many other parameters. See full CRD reference for details.
0 commit comments