app.py

import gradio as gr
import plotly.graph_objects as go
import json
import requests
import os
from PIL import Image
import hopsworks
import joblib
import pandas as pd
import numpy as np

API_KEY = os.getenv("API-KEY-TOMTOM")

# Log into hopsworks
project = hopsworks.login()
fs = project.get_feature_store()
mr = project.get_model_registry()

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


def predict(magnitudeOfDelay, hour, iconCategory, latitude, longitude, month):
    # Create a row from the input
    row = {
        'magnitudeOfDelay': magnitudeOfDelay,
        'hour': hour,
        'iconCategory': iconCategory,
        'latitude': latitude,
        'longitude': longitude,
        'month': month
    }

    # Create a df from the row
    df_row = pd.DataFrame(row, index=[0])

    # change the order to code	hour	iconCategory	latitude	longitude	magnitudeOfDelay	month	duration
    df_row = df_row[['magnitudeOfDelay', 'hour', 'iconCategory', 'latitude', 'longitude', 'month']]

    # make the features lower case
    df_row.columns = df_row.columns.str.lower()
    df_row.columns = df_row.columns.str.replace(' ', '_')

    # Get the prediction
    prediction = model.predict(df_row)[0]
    
    # Remove the log transformation
    prediction = prediction[0]

    return prediction

demo = gr.Interface(
    fn =predict,
    title="Stockholm Incident Prediction",
    description="Predicts the duration of a traffic incident in Stockholm",
    allow_flagging="never",
    inputs=[
        gr.inputs.Slider(0, 60, label="Magnitude of Delay"),
        gr.inputs.Slider(0, 23, label="Hour"),
        gr.inputs.Radio(["Accident", "Construction", "Congestion", "Disabled Vehicle", "Mass Transit", "Miscellaneous", "Other News", "Planned Event", "Road Hazard", "Roadwork", "Traffic Flow", "Weather"], label="Icon Category"),
        gr.inputs.Slider(59.25, 59.40, label="Latitude"),
        gr.inputs.Slider(18.00, 18.16, label="Longitude"),
        gr.inputs.Slider(1, 12, label="Month")
    ],
    outputs=[
        gr.outputs.Textbox(label="Duration")
    ])

demo.launch()