import gradio as gr
import numpy as np
from PIL import Image
import requests

import hopsworks
import joblib

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


mr = project.get_model_registry()
model = mr.get_model("titanic_modal", version=10)
model_dir = model.download()
model = joblib.load(model_dir + "/titanic_model.pkl")

def titanic(age, embarked, fare, parch, pclass, sex, sibsp):
    input_list = []
    input_list.append(age)
    input_list.append(embarked)
    input_list.append(fare)
    input_list.append(parch)
    input_list.append(pclass)
    input_list.append(sex)
    input_list.append(sibsp)
    # 'res' is a list of predictions returned as the label.
    res = model.predict(np.asarray(input_list).reshape(1, -1)) 
    # We add '[0]' to the result of the transformed 'res', because 'res' is a list, and we only want the first element.
    # 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)            
    if res == [1]:
        res = 'survive'
    else:
        res = 'die'
    return res
        
demo = gr.Interface(
    fn=titanic,
    title="Titanic Survivor Predictive Analytics",
    description="Experiment with age/embarked/fare/parch/pclass/sex/sibsp to predict if the passenger survived.",
    allow_flagging="never",
    inputs=[
        gr.inputs.Number(default=2.0, label="age"),
        gr.inputs.Number(default=1.0, label="embarked (0 for S, 1 for C, 2 for Q)"),
        gr.inputs.Number(default=35.0, label="fare"),
        gr.inputs.Number(default=1.0, label="parch"),
        gr.inputs.Number(default=1.0, label="pclass"),
        gr.inputs.Number(default=1.0, label="sex (0 for male, 1 for male)"),
        gr.inputs.Number(default=1.0, label="sibsp")
        ],
    outputs=gr.Textbox())
demo.launch()