File tree 6 files changed +102
-0
lines changed
6 files changed +102
-0
lines changed Original file line number Diff line number Diff line change
1
+ ## Simple chatbot
2
+
3
+ This folder holds the resource definitions to launch a chatbot given a ` MODEL_ENDPOINT ` and ` MODEL_ENDPOINT_BEARER ` .
4
+
5
+ Update the deployment as necessary and
6
+ run this from the root of the repository
7
+
8
+
9
+ ``` bash
10
+ oc apply --kustomize ./chatbot
11
+ ```
12
+
13
+ TODO:
14
+ - get ` MODEL_ENDPOINT ` from configmap or secret. Currently you need to update it in deployment.yaml
15
+
16
+ ### Chatbot
17
+
18
+ The chatbot image is built from
19
+ [ ai-lab-recipes repository chatbot] ( https://github.com/containers/ai-lab-recipes/blob/main/recipes/natural_language_processing/chatbot/app/Containerfile )
20
+ with the below system prompt line from
21
+ [ chatbot_ui.py] ( https://github.com/containers/ai-lab-recipes/blob/main/recipes/natural_language_processing/chatbot/app/chatbot_ui.py )
22
+ commented out, since it's not compatible with vLLM:
23
+
24
+ ``` bash
25
+ prompt = ChatPromptTemplate.from_messages([
26
+ # ("system", "You are world class technical advisor."),
27
+ MessagesPlaceholder(variable_name=" history" ),
28
+ (" user" , " {input}" )
29
+ ])
30
+ ```
Original file line number Diff line number Diff line change
1
+ apiVersion : apps/v1
2
+ kind : Deployment
3
+ metadata :
4
+ name : chatbot
5
+ labels :
6
+ app : chatbot
7
+ spec :
8
+ replicas : 1
9
+ selector :
10
+ matchLabels :
11
+ app : chatbot
12
+ template :
13
+ metadata :
14
+ labels :
15
+ app : chatbot
16
+ spec :
17
+ serviceAccountName : chatbot-sa
18
+ containers :
19
+ - name : chatbot-inference
20
+ image : quay.io/sallyom/chatbot:latest
21
+ env :
22
+ - name : MODEL_ENDPOINT
23
+ # Update this value to the endpoint of a running model server
24
+ value : https://candidatemodel-server-example/v1
25
+ - name : MODEL_ENDPOINT_BEARER
26
+ valueFrom :
27
+ secretKeyRef :
28
+ name : judge-server
29
+ key : api_key
30
+ ports :
31
+ - containerPort : 8501
32
+ securityContext :
33
+ runAsNonRoot : true
Original file line number Diff line number Diff line change
1
+ apiVersion : kustomize.config.k8s.io/v1beta1
2
+ kind : Kustomization
3
+
4
+ resources :
5
+ - deployment.yaml
6
+ - service.yaml
7
+ - route.yaml
8
+ - sa.yaml
Original file line number Diff line number Diff line change
1
+ apiVersion : route.openshift.io/v1
2
+ kind : Route
3
+ metadata :
4
+ name : chatbot
5
+ labels :
6
+ app : chatbot
7
+ spec :
8
+ to :
9
+ kind : Service
10
+ name : chatbot-service
11
+ port :
12
+ targetPort : 8501
13
+ tls :
14
+ termination : edge
Original file line number Diff line number Diff line change
1
+ apiVersion : v1
2
+ kind : ServiceAccount
3
+ metadata :
4
+ name : chatbot-sa
Original file line number Diff line number Diff line change
1
+ apiVersion : v1
2
+ kind : Service
3
+ metadata :
4
+ name : chatbot-service
5
+ labels :
6
+ app : chatbot
7
+ spec :
8
+ selector :
9
+ app : chatbot
10
+ ports :
11
+ - protocol : TCP
12
+ port : 8501
13
+ targetPort : 8501
You can’t perform that action at this time.
0 commit comments