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("iris_model", version=1)
model_dir = model.download()
model = joblib.load(model_dir + "/iris_model.pkl")
print("Model downloaded")


def iris(sepal_length, sepal_width, petal_length, petal_width):
    print("Calling function")
    df = pd.DataFrame([[sepal_length, sepal_width, petal_length, petal_width]],
                      columns=['sepal_length', 'sepal_width', 'petal_length', 'petal_width'])
    print("Predicting")
    print(df)
    res = model.predict(df)
    print(res)
    flower_url = "https://raw.githubusercontent.com/featurestoreorg/serverless-ml-course/main/src/01-module/assets/" + \
                 res[0] + ".png"
    img = Image.open(requests.get(flower_url, stream=True).raw)
    return img


demo = gr.Interface(
    fn=iris,
    title="Iris Flower Predictive Analytics",
    description="Experiment with sepal/petal lengths/widths to predict which flower it is.",
    allow_flagging="never",
    inputs=[
        gr.Number(label="sepal length (cm)"),
        gr.Number(label="sepal width (cm)"),
        gr.Number(label="petal length (cm)"),
        gr.Number(label="petal width (cm)"),
    ],
    outputs=gr.Image(type="pil"))

demo.launch(debug=True)