Fix Helm deployment template, add missing env variables, fix logger (#48)

Signed-off-by: Hung-Han (Henry) Chen <chenhungh@gmail.com>

charts/ialacol/Chart.yaml

 apiVersion: v2
-appVersion: 0.7.1
+appVersion: 0.7.2
 description: A Helm chart for ialacol
 name: ialacol
 type: application
-version: 0.7.2
+version: 0.7.3

charts/ialacol/templates/deployment.yaml

@@ -29,20 +29,40 @@ spec:
             value: {{ (.Values.deployment.env).DEFAULT_MODEL_HG_REPO_ID | quote }}
           - name: DEFAULT_MODEL_FILE
             value: {{ (.Values.deployment.env).DEFAULT_MODEL_FILE | quote }}
-          - name: GPU_LAYERS
-            value: {{ (.Values.deployment.env).GPU_LAYERS | quote }}
+          - name: DOWNLOAD_DEFAULT_MODEL
+            value: {{ (.Values.deployment.env).DOWNLOAD_DEFAULT_MODEL | quote }}
           - name: LOGGING_LEVEL
             value: {{ (.Values.deployment.env).LOGGING_LEVEL | quote }}
           - name: MODELS_FOLDER
             value: {{ (.Values.deployment.env).MODELS_FOLDER | quote }}
           - name: CACHE_FOLDER
             value: {{ (.Values.deployment.env).CACHE_FOLDER | quote }}
-          - name: THREADS
-            value: {{ (.Values.deployment.env).THREADS | quote }}
+          - name: TOP_K
+            value: {{ (.Values.deployment.env).TOP_K | quote }}
+          - name: TOP_P
+            value: {{ (.Values.deployment.env).TOP_P | quote }}
+          - name: TEMPERATURE
+            value: {{ (.Values.deployment.env).TEMPERATURE | quote }}
+          - name: REPETITION_PENALTY
+            value: {{ (.Values.deployment.env).REPETITION_PENALTY | quote }}
+          - name: LAST_N_TOKENS
+            value: {{ (.Values.deployment.env).LAST_N_TOKENS | quote }}
+          - name: SEED
+            value: {{ (.Values.deployment.env).SEED | quote }}
           - name: BATCH_SIZE
             value: {{ (.Values.deployment.env).BATCH_SIZE | quote }}
+          - name: THREADS
+            value: {{ (.Values.deployment.env).THREADS | quote }}
+          - name: MAX_TOKENS
+            value: {{ (.Values.deployment.env).MAX_TOKENS | quote }}
+          - name: STOP
+            value: {{ (.Values.deployment.env).STOP | quote }}
           - name: CONTEXT_LENGTH
             value: {{ (.Values.deployment.env).CONTEXT_LENGTH | quote }}
+          - name: GPU_LAYERS
+            value: {{ (.Values.deployment.env).GPU_LAYERS | quote }}
+          - name: MODE_TYPE
+            value: {{ (.Values.deployment.env).MODE_TYPE | quote }}
           volumeMounts:
           - mountPath: {{ .Values.cacheMountPath }}
             name: cache

get_config.py

@@ -7,7 +7,7 @@ from get_default_thread import get_default_thread
 
 LOGGING_LEVEL = get_env("LOGGING_LEVEL", "INFO")
 
-log = logging.getLogger(__name__)
+log = logging.getLogger("uvicorn")
 try:
     log.setLevel(LOGGING_LEVEL)
 except ValueError:

main.py

@@ -32,7 +32,7 @@ LOGGING_LEVEL = get_env("LOGGING_LEVEL", "INFO")
 MODELS_FOLDER = get_env("MODELS_FOLDER", "models")
 CACHE_FOLDER = get_env("MODELS_FOLDER", "cache")
 
-log = logging.getLogger(__name__)
+log = logging.getLogger("uvicorn")
 
 log.info("DEFAULT_MODEL_HG_REPO_ID: %s", DEFAULT_MODEL_HG_REPO_ID)
 log.info("DEFAULT_MODEL_FILE: %s", DEFAULT_MODEL_FILE)