|
import gradio as gr |
|
import numpy as np |
|
from PIL import Image |
|
import requests |
|
from feature_engineering import feat_eng |
|
import hopsworks |
|
import joblib |
|
import pandas as pd |
|
|
|
project = hopsworks.login() |
|
fs = project.get_feature_store() |
|
sharebool = True |
|
|
|
mr = project.get_model_registry() |
|
model = mr.get_model("titanic_modal_simple_classifier", version=1) |
|
model_dir = model.download() |
|
model = joblib.load(model_dir + "/titanic_model.pkl") |
|
leo_url = "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAkGBxAQDw8PEA4NDw0PDw0PDQ8NDw8NDQ0NFREWFhURFRUYHSggGBolGxUVITEhJSkrLi4uFx8zOD8tNygtLisBCgoKDg0OFxAQFy0dHR8tLS0tLS0tLS0tLS0tLS0tLS0tLS0tOC4tLS0tLSstLS0tKy0rLSstLS0rLSstNy0tK//AABEIAMgA/AMBIgACEQEDEQH/xAAbAAEAAgMBAQAAAAAAAAAAAAAAAwQBAgUGB//EAC4QAAICAQIFBAICAQUBAAAAAAABAgMRBCESMUFTkgUUUWEicRORgQYyUnLRFf/EABkBAQADAQEAAAAAAAAAAAAAAAABAwQCBf/EACIRAQEAAwACAgIDAQAAAAAAAAABAgMREiEEMSIyE0FRQv/aAAwDAQACEQMRAD8A+J+6s7lnnIe6s7lnnIhBIm91Z3LPOQ91Z3LPOREBwS+6s7lnnIz7mzuWeciAySJvc2dyzzkY91Z3LPOREYIE/ubO5Z5yMe6s7lnnIhBAm91Z3LPOQ91Z3LPORCAJvdWdyzzkPdWdyzzkQgnom91Z3LPOX/pn3Vncs85EAIE/ubO5Z5yMe6s7lnnIiAEvurO5Z5yHubO5Z5yIgBL7mzuWech7mzuWeciIAS+5s7lnnIe5s7lnnIiAEvurO5Z5yHurO5Z5yIgBL7qzuWech7qzuWeciIwBN7qzuWech7qzuWeciIICX3Vncs85D3Vncs85EZqAAAAAE9AADoAAgAAAAAGRg2hHcuxo2O8cOuMs5ioYMFyNO+CLUU8LwRcbCZyoQ0ZSDZy6YAASAAAAYAyDBkAYAAyDAAyYAAAAAAAAAAAAAAAAMpEkKyZjai3iXS15aOrZViBr6bpupNreWDXhjzFj2Z9yc+H+7JLqq+JJmjwupur1w4OXUc62vBC0Xb2nyK3CU5Y+1+OXpGkGS4I5HNjuVqADlIAAAAAAAAAAAAAAAAAAAAAAAAAZigJaoZZ1KNDFrmylRWzoV1Sxu9ka9eE4y7M2vE63hPOCxU+PmUrXudL0mvMW2v0XY+7xTl9dVrNLgOqLjy/WDtTojJblZ+m/Emkzr+LrmbHHelILKMHpIen4XPP7ItXpEoSeN8Mry0O8d0eXmiKRNZnLIJMyZTjZi1ABW7AAAAAAAAAAAAAAAAAAAAAAAADaBqZRM+x0dNLdHXjhx/wcOg6entwjfr+mHbOVSsg0zoenaiS/ElxF80iXTOuPxllmOPL1xln2LVLk3uv8otwI67Y45okUi6M9ZnIqah5TXR7FiyRztbqFBZfIZX0Y4230qXaWMVKTX4pN7nnJc2dn1P1GMocEd84y+RxmeZvs76elplk9sAAoXgAAAAAAAAAAAAAAAAAAAAAAZwBgFmOjnjLjJL7RC4HfhUeUb1Wtfo6NE+XwcrBd008ov1Ze+VTtxnOurasrKIq6Gzai3ozo0Vpo2SdYreIKNK882joVppc8msIYZKyznIruXWsmcX1tZgtuUv6OtZPBzPUN4sqzvca713mUeekaqDfRl2FCb3L2ioip/XQx46vKt2W6Yxw2jB2PUPTXxNw5c8HLnTJc0/6K89WWN+neGzHKekYM8L+GGivixgAEAAAAAAAAAAAAAAAyotksNO30Opjb9ItkRrBPprVGSfx8rJNX6bJ/om/+bLbCbe3Q0468/vinLbh9dWLNa5RzzOXNneXp/DW01mWDiWV4Zbt7Iq1XH3xC0b6bmacRPTX1M+Pu+l+V5PazFnQ0upwt2UaoZRvwY6mqXjJlJXYhqo/Ilq10OTVBtnRqpWCyW1VcZGlkpSZFcsRbfwy24kOpr4otfRzl9E+3Gk8Iu+l/lu+nI513wdn0SCVWX1bKNP7r9vJh1Lj8je6lNZ+smsLFl/I1mp/HgNvZxmkqCjTriTxs0Rep+lOX5QW/VFvSLZZL/Ec/x45Y8p/Lljl2PDzg1s00/vY0PcW6WFm0op/fU5eq/wBOp5cJNfTWTHn8PKfr7bMPl4X79PNgl1GnlB8Mk0/siMlll5WqWX3AAyQlgAAAAANq45aRqWNEvzR1hO5SIyvJa6+mpSikkv20i3HTr/iawjsi5RHY9fHCPJzztaQrSJq9mZupytufQrwk1s00yxV9rU9zi+o6TDbS2Z1VITrUlhleePks15eLyMoYZZpRa9U0nBNfDWxjTVZMeOHLW3LZLjKRybJstfxbGih9FvFXW9KLlbK9aLFZ3FVb4NeElUSRRCHl9dXiT/bJdHrOGHAvks+rJN/9c5ONLZmS3wy614yZ48rrKeTGlk23nfHIrUS25l3Sx6l+Pv2qynj2LtMixGwr1ksDTiy5e1qLJFNGla2MTrysZxktlV8U/WNIrIf7fyXJrn+jx8ljKPWaz+aCfD+S3/Z5a7PE8ppt5aZ5vzJ+XePS+Jfx4jBkGJsYAAAAAZSydX03SNfk+vL9EPp2ly1J8un7O/RVt8G742n/AKrH8jdyeMZhXlJFyuOEiOKwTI3yPP6yayWTPEYcgiRXccPBsmby3REQlU9WjlRfwQaaOUXNXFODWfs5+ls6fBmz/Zpw/VP/ACYeGWKsMq6hZw+vURta3XLqILzoMwhgxTq4uO+xo9TH5OnNW1Ej1N/AiCWrKtlvFzYtJFTUSby31OfaXtW9jnzZk2tepvVZg6NFnI5Ce5bomdas07cOu5U9sliCOdpZlxW4Rtl9MGU9rf8AKoognqm+WxWlZkwnuT1HI6NUsrc89/qGrE1LGzR3FPEcnM9ahxVZ6xK/kYzLBZ8e+Ox54GAeS9UAAA6Gl03LbLZU08MtHo9DTGKT5v8Ao1/G199s3yNnjE+l02Es/wBFlmqkGz0pOPNt6zBmZTNUQ2yHXMiXiNuIrRkLdRGKbZz5O5jallP7NG2QV2KayuRuOnOM2SWGn8HIqf5NIu6uTSZy4TxIy7b+UadWPZXSrkYcuGX0yBz6oy7OJbjp4pVLEvo01M1+iJ2Z26mykmmmOpmKxpdT0e5JOaZzoLDynsSrUJ/QmX+puP8AjXVvY50y/dLKOfPcz7vtfq+iBcpRXqgWYDVKbKt1SwTcZWiyRM1ystxT8QjPch4jHETa48V2y38cGleJJxlyawyrJm9csEyo8VD1L07+P8k8xf8AaOcetpUZpxmsxax+n8nD1vprrbfC3B7prfC+zNv0c94/TXo39/HL7c0GZIwYmt09Fp8bnSi30LENOksYwkYmkj19eHhHl55+d6zXYyeJTc8E1VvyWRTYnkyvYyac1gqXyFMRSOf6pb+OPssubOb6hl/oz7bzGtWnH8mNFqnHbLSbO3GyPCnxLD+zzEWSKf2zPr38X7dMyrpay7Odyjk1laaqRzln3LrrDX4zi/W9jXiwRxlsYyWdV+KTPUJs1gZJlGqlhm7SZDPmbpnPXVjW2bSIKzN9mTSEinPLuSzHHkWokkSKDJYl+KrJPBm+SKLN8lsU1tkxkwxElHG6NomqJa4kxzVilbFqMsrDIqYk3Ci+T0pt9vMepaWUJtuOIt7Nciiey1dHHBx69Mnkr6JRlwyTTPN+Rp8L2f29HRu85y/09TPi+COVbfMt2SIZM9OvOlQfxYMqJs2YcznvE9ORDdNEV1jKV8ngryz5FuGvqxZainq7srCK07WRtmLZu7ONmGnntg2TNTKMy9nJgAnosweUFI0pkZsXUvmXpVz23c8G38pDnJHJjz4ePU1kjX+TYiyYK7nXUxZbMZAK3SeuwnhYUUyWEy7HYrywX4s3yVIS+yWMjTM4ouKbJtEjUiSLO1diRFilFdSLFUkdRVkt1E2CrXYWOMvxqmpkiLUaaMnlpN4NozMWT3/wTZL9ols+lN6uJFZqV0AK/KtE1xmLybxpfygDqTqu3ivqIJN7nN1ElvuYBn33katE6oswAea3AAAAADaMjaUwDryRxqpGGwCOpAYBAAADITAA2U2TQ1L64AOplY5uMq1XbF9VklbXyv7ANeGdsZ8tchxfaN65sAslV5Y+k8GWlNfIBdKz2NlfH5E7V8oA76jwf//Z" |
|
rose_url = "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSGoi8okN1Fw6tYE7k-0H-wnMabl1e3NBNPpQ&usqp=CAU" |
|
|
|
|
|
|
|
def titanic(pclass, name, sex, age, sibsp, parch, ticket, fare, cabin, embarked): |
|
df_pre = pd.DataFrame({"PassengerId":[-1], "Pclass": [pclass], "Name": [name], "Sex": [sex], "Age": [age], "SibSp": [sibsp], "Parch": [parch], "Ticket": [ticket], "Fare": [fare], "Cabin": [cabin], "Embarked": [embarked]}) |
|
df_post = feat_eng(df_pre) |
|
|
|
res = model.predict(df_post)[0] |
|
|
|
|
|
|
|
img = Image.open(leo_url) if res == 0 else Image.open(rose_url) |
|
return img |
|
|
|
demo = gr.Interface( |
|
fn=titanic, |
|
title="Titanic Survival Predictive Analytics", |
|
description="Experiment with Titanic Passenger data to predict survival", |
|
allow_flagging="never", |
|
inputs=[ |
|
gr.inputs.Number(default=1.0, label="pclass, [1,2,3]"), |
|
gr.inputs.Textbox(default="Mr. Anton", label="name"), |
|
gr.inputs.Textbox(default="male", label="sex, male or female"), |
|
gr.inputs.Number(default=25, label="age"), |
|
gr.inputs.Number(default=2, label="sibsb"), |
|
gr.inputs.Number(default=2, label="parch"), |
|
gr.inputs.Textbox(default="blabla", label="Ticket"), |
|
gr.inputs.Number(default=200, label="Fare"), |
|
gr.inputs.Textbox(default="A123", label="Cabin"), |
|
gr.inputs.Textbox(default="S", label="Embarked: [S, C, Q]") |
|
], |
|
outputs=gr.Image(type="pil")) |
|
|
|
demo.launch() |
