implement prediction endpoint and update Dockerfile and requirements

Dockerfile

@@ -1,5 +1,12 @@
 FROM python:3.9
 
+ENV MLFLOW_EXPERIMENT_NAME=$MLFLOW_EXPERIMENT_NAME
+ENV MLFLOW_TRACKING_URI=$MLFLOW_TRACKING_URI
+ENV AWS_ACCESS_KEY_ID=$AWS_ACCESS_KEY_ID
+ENV AWS_SECRET_ACCESS_KEY=$AWS_SECRET_ACCESS_KEY
+ENV BACKEND_STORE_URI=$BACKEND_STORE_URI
+ENV ARTIFACT_ROOT=$ARTIFACT_ROOT
+
 RUN useradd -m -u 1000 user
 USER user
 ENV PATH="/home/user/.local/bin:$PATH"

README.md

@@ -9,4 +9,3 @@ license: mit
 short_description: Web service for Getaround app to deliver car rental pricing
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -1,7 +1,41 @@
 from fastapi import FastAPI
+from pydantic import BaseModel
+import mlflow.pyfunc
+import pandas as pd
 
 app = FastAPI()
 
+
 @app.get("/")
 def greet_json():
     return {"Hello": "World!"}
+
+
+class PredictionFeatures(BaseModel):
+    YearsExperience: float
+
+
+#### SOME CODE ####
+###################
+
+
+@app.post("/predict", tags=["Machine Learning"])
+async def predict(predictionFeatures: PredictionFeatures):
+    """
+    Prediction of salary for a given year of experience!
+    """
+    # Read data
+    years_experience = pd.DataFrame(
+        {"YearsExperience": [predictionFeatures.YearsExperience]}
+    )
+
+    # Log model from mlflow
+    logged_model = "runs:/c09d09ef14e546b08f2f339d2c966da6/salary_estimator"  # REPLACE WITH YOUR OWN RUN ID
+
+    # Load model as a PyFuncModel.
+    loaded_model = mlflow.pyfunc.load_model(logged_model)
+    prediction = loaded_model.predict(years_experience)
+
+    # Format response
+    response = {"prediction": prediction.tolist()[0]}
+    return response

requirements.txt

@@ -1,2 +1,7 @@
 fastapi
 uvicorn[standard]
+pydantic
+mlflow
+pandas
+
+