fix typo

module-5/Dockerfile

@@ -18,14 +18,12 @@ CMD [ "bash" ]
 FROM base AS app-streamlit
 CMD streamlit run --server.address 0.0.0.0 --server.port 8080 serving/ui_app.py
 
-
 # Fast API docker image
 FROM base AS app-fastapi
 CMD uvicorn --host 0.0.0.0 --port 8080 --workers 4 serving.fast_api:app 
 
 FROM base AS app-pytriton
 CMD python serving/pytriton_serving.py
 
-
 FROM base AS app-kserve
 ENTRYPOINT ["python", "serving/kserve_api.py"]

module-5/README.md

@@ -99,6 +99,12 @@ Deploy custom model
 kubectl create -f ./k8s/kserve-inferenceserver.yaml
 ```
 
+Port forward via istio
+
+```
+kubectl port-forward --namespace istio-system svc/istio-ingressgateway 8080:80
+```
+
 Call API 
 
 ```
@@ -149,7 +155,7 @@ Deploy
 Run K8S with GPUs
 
 ```
- curl -LO https://storage.googleapis.com/minikube/releases/latest/minikube_latest_amd64.deb
+curl -LO https://storage.googleapis.com/minikube/releases/latest/minikube_latest_amd64.deb
 sudo dpkg -i minikube_latest_amd64.deb
 minikube start --driver docker --container-runtime docker --gpus all
 ```

module-5/k8s/app-streamlit.yaml

@@ -4,7 +4,7 @@ kind: Deployment
 metadata:
   name: app-streamlit
 spec:
-  replicas: 2
+  replicas: 1
   selector:
     matchLabels:
       app: app-streamlit

module-5/serving/kserve_api.py

@@ -1,4 +1,3 @@
-import json
 from serving.predictor import Predictor
 from typing import Dict
 from kserve import Model, ModelServer
@@ -17,8 +16,6 @@ def load(self):
     def predict(self, payload: Dict, headers: Dict[str, str] = None) -> Dict:
         print(payload)
         print(type(payload))
-        # json_payload = json.loads(payload.decode("utf-8"))
-        # instances = json_payload["instances"]
         instances = payload["instances"]
         predictions = self.predictor.predict(instances)
         return {"predictions": predictions.tolist()}

module-5/serving/pytriton_client.py

@@ -1,4 +1,3 @@
-import logging
 import numpy as np
 from pytriton.client import ModelClient