Create app.py

app.py

@@ -0,0 +1,29 @@
+# -*- coding: utf-8 -*-
+
+import pandas as pd
+from pycaret.regression import load_model, predict_model
+from fastapi import FastAPI
+import uvicorn
+from pydantic import create_model
+
+# Create the app
+app = FastAPI()
+
+# Load trained Pipeline
+model = load_model("lr_api")
+
+# Create input/output pydantic models
+input_model = create_model("lr_api_input", **{'rownames': 1030, 'year': 1994, 'violent': 304.5, 'murder': 2.9000000953674316, 'prisoners': 152, 'afam': 1.769081950187683, 'cauc': 70.66014862060547, 'male': 18.20832061767578, 'population': 1.9304360151290894, 'income': 12036.8603515625, 'density': 0.023493800312280655, 'state': 'Utah', 'law': 'yes'})
+output_model = create_model("lr_api_output", prediction=63.6)
+
+
+# Define predict function
+@app.post("/predict", response_model=output_model)
+def predict(data: input_model):
+    data = pd.DataFrame([data.dict()])
+    predictions = predict_model(model, data=data)
+    return {"prediction": predictions["prediction_label"].iloc[0]}
+
+
+#if __name__ == "__main__":
+#    uvicorn.run(app, host="127.0.0.1", port=8000)