app.py

import gradio as gr
from PIL import Image
import requests
import hopsworks
import joblib
import pandas as pd

project = hopsworks.login()
fs = project.get_feature_store()


mr = project.get_model_registry()
model = mr.get_model("wine_model", version=1)
model_dir = model.download()
model = joblib.load(model_dir + "/wine_model.pkl")
print("Model downloaded")

def wine(type, volatile_acidity, citric_acid, chlorides, density, sulphates, alcohol):
    print("Calling function")
    df = pd.DataFrame([[type, volatile_acidity, citric_acid, chlorides, density, sulphates, alcohol]], 
                      columns=['type', 'volatile_acidity', 'citric_acid', 'chlorides', 'density', 'sulphates', 'alcohol'])
    print("Predicting")
    print(df)
    # 'res' is a list of predictions returned as the label.
    res = model.predict(df) 
    # We add '[0]' to the result of the transformed 'res', because 'res' is a list, and we only want 
    # the first element.
    print(f"Res: {res[0]}")
         
    return res[0]
        
demo = gr.Interface(
    fn=wine,
    title="Wine Quality Predictive Analytics",
    description="Experiment with type (red/white), volatile acidity, citric acid, chlorides, density, sulphates, alcohol, quality to predict the wine's quality.",
    allow_flagging="never",
    inputs=[
        gr.Number(value=1.0, label="wine type (red = 1, white = 0)"),
        gr.Number(value=1.0, label="Volatile acidity"),
        gr.Number(value=1.0, label="citric_acid"),
        gr.Number(value=1.0, label="chlorides"),
        gr.Number(value=1.0, label="density"),
        gr.Number(value=1.0, label='sulphates'),
        gr.Number(value=1.0, label='alcohol'),
        ],
    outputs=gr.Number(label="quality"))

demo.launch(debug=True)