added signature in mlflow example input

Dockerfile

@@ -1,8 +1,18 @@
 FROM python:3.10-slim
 
-RUN apt update -y && apt install awscli -y
+RUN apt update -y && apt install awscli -y git
+
+# Set the Git executable path for GitPython
+ENV GIT_PYTHON_GIT_EXECUTABLE=/usr/bin/git
+
 WORKDIR /app
 
+# Create a virtual environment
+RUN python3 -m venv /opt/venv
+
+# Activate the virtual environment
+ENV PATH="/opt/venv/bin:$PATH"
+
 COPY . /app
 RUN pip install -r requirements.txt
 

src/mlProject/components/model_evaluation.py

@@ -9,6 +9,8 @@
 from mlProject.entity.config_entity import ModelEvaluationConfig
 from pathlib import Path
 from mlProject.utils.common import save_json
+from mlflow.models import ModelSignature
+from mlflow.types import Schema, ColSpec
 
 class ModelEvaluation:
     def __init__(self, config: ModelEvaluationConfig):
@@ -28,6 +30,27 @@ def log_into_mlflow(self):
 
         test_x = test_data.drop([self.config.target_column], axis=1)
         test_y = test_data[[self.config.target_column]]
+
+        # Create an input example (this should match your model's input shape)
+        input_example = test_data.drop([self.config.target_column], axis=1).iloc[0:1]  # Take the first row as an example
+
+        # Define the model signature
+        input_schema = Schema([
+            ColSpec("float", "fixed acidity"),  # Replace with your actual feature names and types
+            ColSpec("float", "volatile acidity"),
+            ColSpec("float", "citric acid"),
+            ColSpec("float", "residual sugar"),
+            ColSpec("float", "chlorides"),
+            ColSpec("float", "free sulfur dioxide"),
+            ColSpec("float", "total sulfur dioxide"),
+            ColSpec("float", "density"),
+            ColSpec("float", "pH"),
+            ColSpec("float", "sulphates"),
+            ColSpec("float", "alcohol"),
+        ])
+
+        # Create the model signature
+        signature = ModelSignature(inputs=input_schema)
 
         mlflow.set_registry_uri(self.config.mlflow_uri)
         tracking_url_type_store = urlparse(mlflow.get_tracking_uri()).scheme
@@ -55,6 +78,15 @@ def log_into_mlflow(self):
                 # There are other ways to use the Model Registry, which depends on the use case,
                 # Please refer to the doc for more information
                 # https://mlflow.org/docs/latest/model-registry.html#api-workflow
-                mlflow.sklearn.log_model(model, "model", registered_model_name = "ElasticnetModel")
+                mlflow.sklearn.log_model(model,
+                                         "model",
+                                         registered_model_name = "ElasticnetModel",
+                                         signature=signature,
+                                         input_example=input_example.to_dict(orient="records")[0]
+                                         )
             else:
-                mlflow.sklearn.log_model(model, "model")
+                mlflow.sklearn.log_model(model,
+                                         "model",
+                                         signature=signature,
+                                         input_example=input_example.to_dict(orient="records")[0]
+                                         )