[pvc-docs] fixing documention for storage PVC configuration (#400)

kserve · Oct 12, 2024 · 94c12e8 · 94c12e8
1 parent 2fb5470
commit 94c12e8
Show file tree

Hide file tree

Showing 3 changed files with 12 additions and 5 deletions.
diff --git a/docs/modelserving/storage/pvc/pvc.md b/docs/modelserving/storage/pvc/pvc.md
@@ -103,7 +103,7 @@ Update the ${PVC_NAME} to the created PVC name and create the InferenceService w
         model:
           modelFormat:
             name: sklearn
-          storageUri: "pvc://${PVC_NAME}/model.joblib"
+          storageUri: "pvc://${PVC_NAME}/${MODEL_NAME}/"
     ```
 
 === "Old Schema"
@@ -116,16 +116,23 @@ Update the ${PVC_NAME} to the created PVC name and create the InferenceService w
     spec:
       predictor:
         sklearn:
-          storageUri: "pvc://${PVC_NAME}/model.joblib"
+          storageUri: "pvc://${PVC_NAME}/${MODEL_NAME}/"
     ```
 
-Apply the `autoscale-gpu.yaml`.
+Apply the `sklearn-pvc.yaml`.
 
 === "kubectl"
 ```bash
 kubectl apply -f sklearn-pvc.yaml
 ```
 
+Note that inside the folder `${PVC_NAME}/${MODEL_NAME}/` you should have your
+model `model.joblib`.
+
+Note also that `${MODEL_NAME}` is just a folder, but a good convention to keep
+the same name.
+
+
 ## Run a prediction
 
 Now, the ingress can be accessed at `${INGRESS_HOST}:${INGRESS_PORT}` or follow [this instruction](../../../get_started/first_isvc.md#4-determine-the-ingress-ip-and-ports)

diff --git a/docs/modelserving/storage/pvc/sklearn-pvc-new.yaml b/docs/modelserving/storage/pvc/sklearn-pvc-new.yaml
@@ -7,4 +7,4 @@ spec:
     model:
       modelFormat:
         name: sklearn
-      storageUri: "pvc://${PVC_NAME}/model.joblib"
+      storageUri: "pvc://${PVC_NAME}/sklearn-pvc"
diff --git a/docs/modelserving/storage/pvc/sklearn-pvc.yaml b/docs/modelserving/storage/pvc/sklearn-pvc.yaml
@@ -5,4 +5,4 @@ metadata:
 spec:
   predictor:
     sklearn:
-      storageUri: "pvc://${PVC_NAME}/model.joblib"
+      storageUri: "pvc://${PVC_NAME}/sklearn-pvc"