mlflow attempt task 2 prediction registry

Jenkinsfile_pred_reg

@@ -0,0 +1,28 @@
+pipeline {
+    agent { 
+        dockerfile  true
+    }
+    parameters {
+        string(
+			defaultValue: 'input_example.json',
+			description: 'Input file name',
+			name: 'INPUT_FILE_NAME',
+			trim: false
+        )
+	}
+
+    stages {
+        stage('Stage 1') {
+            steps {
+                echo 'Hello world!' 
+            }
+        }
+
+        stage('Prediction') {
+            steps {
+                    sh 'python3 predict_registry.py $INPUT_FILE_NAME'                
+            }
+        }
+    }
+}
+

predict_registry.py

@@ -0,0 +1,32 @@
+import json
+import mlflow
+import numpy as np
+import sys
+
+mlflow.set_tracking_uri("http://172.17.0.1:5000")
+client =mlflow.tracking .MlflowClient()
+model_version = 14
+model_name = "s449288"
+
+experiment = client.get_latest_versions(model_name, stages=None)
+print(experiment)
+print(experiment[0].source)
+model = mlflow.pyfunc.load_model(
+    model_uri=f"models:/{model_name}/{model_version}"
+)
+
+with open(f'{experiment[0].source}/{(sys.argv[1:])[0]}', 'r') as file:
+    json_data = json.load(file)
+    print(f"Prediction: {model.predict(np.array([json_data['inputs']]))}")
+
+
+'''
+PATH = "mlruns/14/80fe21a0804844088147d15a3cebb3e5/artifacts/lego-model"
+
+model = mlflow.pyfunc.load_model(PATH)
+
+with open(f'{PATH}/{(sys.argv[1:])[0]}', 'r') as file:
+    json_data = json.load(file)
+    print(f"Prediction: {model.predict(np.array([json_data['inputs']]))}")
+'''
+