refactored GaqGen

FaqGen/benchmark/performance/helm_charts/customize.yaml

@@ -1,13 +1,8 @@
 # Copyright (C) 2024 Intel Corporation
 # SPDX-License-Identifier: Apache-2.0
 
-HUGGINGFACEHUB_API_TOKEN: ${HF_TOKEN}
-
 podSpecs:
   - name: faq-mega-server-deploy
-    spec:
-      image_name: opea/chatqna
-      image_tag: latest
     replicas: 2
     resources:
       limits:
@@ -19,16 +14,10 @@ podSpecs:
 
 
   - name: faq-tgi-deploy
-    spec:
-      image_name: ghcr.io/huggingface/tgi-gaudi
-      image_tag: 2.0.5
     replicas: 7
     resources:
       limits:
         habana.ai/gaudi: 1
 
   - name: faq-micro-deploy
-    spec:
-      image_name: opea/llm-faqgen-tgi
-      image_tag: latest
     replicas: 1

FaqGen/benchmark/performance/helm_charts/values.yaml

@@ -7,20 +7,24 @@ config:
   LLM_MODEL_ID: meta-llama/Meta-Llama-3-8B-Instruct
   CONFIG_MAP_NAME: faq-config
   NODE_SELECTOR: faq-opea
+  HUGGINGFACEHUB_API_TOKEN: ${HF_TOKEN}
 
-deployments:
+microservices:
   - name: faq-mega-server-deploy
-    spec:
+    image: opea/chatqna:latest
+    replicas: 1
     ports:
       - containerPort: 7777
 
   - name: faq-micro-deploy
-    spec:
+    image: opea/llm-faqgen-tgi:latest
+    replicas: 1
     ports:
       - containerPort: 9000
 
   - name: faq-tgi-deploy
-    spec:
+    image: ghcr.io/huggingface/tgi-gaudi:2.0.5
+    replicas: 1
     ports:
       - containerPort: 80
     resources: