Adjust to zenml-io/zenml#3096

Author: AlexejPenner

template/quickstart.ipynb

@@ -982,8 +982,8 @@
     "@pipeline\n",
     "def inference(preprocess_pipeline_id: UUID):\n",
     "    \"\"\"Model batch inference pipeline\"\"\"\n",
-    "    # random_state = client.get_artifact_version(name_id_or_prefix=preprocess_pipeline_id).metadata[\"random_state\"].value\n",
-    "    # target = client.get_artifact_version(name_id_or_prefix=preprocess_pipeline_id).run_metadata['target'].value\n",
+    "    # random_state = client.get_artifact_version(name_id_or_prefix=preprocess_pipeline_id).metadata[\"random_state\"]\n",
+    "    # target = client.get_artifact_version(name_id_or_prefix=preprocess_pipeline_id).run_metadata['target']\n",
     "    random_state = 42\n",
     "    target = \"target\"\n",
     "\n",

template/run.py

@@ -207,8 +207,8 @@ def main(
 
         # Use the metadata of feature engineering pipeline artifact
         #  to get the random state and target column
-        random_state = preprocess_pipeline_artifact.run_metadata["random_state"].value
-        target = preprocess_pipeline_artifact.run_metadata["target"].value
+        random_state = preprocess_pipeline_artifact.run_metadata["random_state"]
+        target = preprocess_pipeline_artifact.run_metadata["target"]
         run_args_inference["random_state"] = random_state
         run_args_inference["target"] = target
 

template/steps/model_promoter.py

@@ -48,7 +48,6 @@ def model_promoter(accuracy: float, stage: str = "production") -> bool:
             prod_accuracy = (
                 stage_model.get_artifact("sklearn_classifier")
                 .run_metadata["test_accuracy"]
-                .value
             )
             if float(accuracy) > float(prod_accuracy):
                 # If current model has better metrics, we promote it