app.py

import pickle
import pandas as pd
import shap
from shap.plots._force_matplotlib import draw_additive_plot
import gradio as gr
import numpy as np
import matplotlib.pyplot as plt

# load the model from disk
loaded_model = pickle.load(open("heart_xgb.pkl", 'rb'))

# Setup SHAP
explainer = shap.Explainer(loaded_model) # PLEASE DO NOT CHANGE THIS.

gender = {"Female":0,"Male":1}
chest = {"typical angina":1,"atypical angina":2,"non-anginal pain":3,"asymptomatic":4}
choice = {"True":1, "False":0}
rest = {"Probable or Definite Left Ventricular Hypertrophy by Estes' Criteria":2, "Having ST - T Wave Abnormality":1,"Normal Value":0}
ex = {"yes":1, "no":0}
sl = {"upsloping":2, "flat":1,"downsloping":0}
th = {"Reversible Defect":3, "Normal Blood Flow":2, "Fixed Defect":1}

# Create the main function for server
def main_func(age, sex, cp, trtbps, chol, fbs, restecg, thalachh, exng, oldpeak, slp, caa, thall):
    new_row = pd.DataFrame.from_dict({'age':age,'sex':gender[sex],
              'cp':chest[cp],'trtbps':trtbps,'chol':chol, 'fbs':choice[fbs], 'restecg':rest[restecg],
              'thalachh':thalachh, 'exng':ex[exng], 'oldpeak':oldpeak, 'slp':sl[slp], 'caa':caa, 'thall':th[thall]}, orient = 'index').transpose()
    
    prob = loaded_model.predict_proba(new_row)
    
    shap_values = explainer(new_row)
    # plot = shap.force_plot(shap_values[0], matplotlib=True, figsize=(30,30), show=False)
    # plot = shap.plots.waterfall(shap_values[0], max_display=6, show=False)
    plot = shap.plots.bar(shap_values[0], max_display=6, order=shap.Explanation.abs, show_data='auto', show=False)

    plt.tight_layout()
    local_plot = plt.gcf()
    plt.close()
    
    return {"Low Heart Attack Chance": float(prob[0][0]), "High Heart Attack Chance": 1-float(prob[0][0])}, local_plot

# Create the UI
title = "**Heart Attack Predictor & Interpreter** 🪐"
description1 = """
This app takes info from subjects and predicts their heart attack likelihood. Do not use for medical diagnosis.   
"""

description2 = """
To use the app, click on one of the examples, or adjust the values of the factors, and click on Analyze. 

Data Dictionary:

Age : Age of the patient

Sex : Sex of the patient

cp : Chest Pain type 

trtbps: Resting Blood Pressure (in mm Hg)

chol : Cholestoral in mg/dl Fetched via BMI Sensor

fbs : (fasting blood sugar > 120 mg/dl)

rest_ecg : Resting Electrocardiographic Results

thalachh : Maximum Heart Rate Achieved

exng: Exercise Induced Angina 

oldpeak : Short-Term Depression Induced by Exercise Relative to Rest

slp : Slope of the Peak Exercise Short-Term Segment

caa : Number of Major Vessels (0-3)

thall : Thalassemia
""" 

with gr.Blocks(title=title) as demo:
    gr.Markdown(f"## {title}")
#    gr.Markdown("""![marketing](file/marketing.jpg)""")
    gr.Markdown(description1)
    gr.Markdown("""---""")
    gr.Markdown(description2)
    gr.Markdown("""---""")
    
    with gr.Row():
        with gr.Column():
            age = gr.Slider(label="age score", minimum=0, maximum=90, value=40, step=1)
            sex = gr.Dropdown(label="sex score", choices =["Female","Male"])
            cp = gr.Radio(label="cp score", choices = ["typical angina", "atypical angina", "non-anginal pain", "asymptomatic"])
            trtbps = gr.Slider(label="trtbps Score", minimum=90, maximum=200, value=90, step=1)
            chol = gr.Slider(label="chol Score", minimum=120, maximum=570, value=120, step=1)
            fbs = gr.Radio(label="fbs Score", choices = ["True", "False"])
            restecg = gr.Dropdown(label="restecg Score", choices = ["Probable or Definite Left Ventricular Hypertrophy by Estes' Criteria", "Having ST - T Wave Abnormality","Normal Value"])
        with gr.Column():
            thalachh = gr.Slider(label="thalachh Score", minimum=70, maximum=210, value=70, step=1) 
            exng = gr.Slider(label="exng Score", choices = ["yes","no"])
            oldpeak = gr.Slider(label="oldpeak Score", minimum=0.0, maximum=6.5, value=0, step=0.1)
            slp = gr.Radio(label="slp Score", choices = ["upsloping", "flat", "downsloping"])
            caa = gr.Slider(label="caa Score", minimum=0, maximum=3, value=0, step=1)
            thall = gr.Slider(label="thall Score", choices = ["Reversible Defect", "Normal Blood Flow", "Fixed Defect"])
    
    with gr.Row(): 
        with gr.Column():
            submit_btn = gr.Button("Analyze")
        
        with gr.Column(visible=True) as output_col:
            label = gr.Label(label = "Predicted Label")
            local_plot = gr.Plot(label = 'Shap:')

    submit_btn.click(
        main_func,
        [age, sex, cp, trtbps, chol, fbs, restecg, thalachh, exng, oldpeak, slp, caa, thall],
        [label,local_plot], api_name="Employee_Turnover"
    )
    
    gr.Markdown("### Click on any of the examples below to see how it works:")
    gr.Examples([[24,"Male","typical angina",150,300,"True","Probable or Definite Left Ventricular Hypertrophy by Estes' Criteria",90,"yes",1,"flat",3,"Reversible Defect"], [20,"Female","asymptomatic",50,400,"True","Having ST - T Wave Abnormality",150,"no",3.5,"upsloping",3,"Fixed Defect"]], [age, sex, cp, trtbps, chol, fbs, restecg, thalachh, exng, oldpeak, slp, caa, thall], [label,local_plot], main_func, cache_examples=True)

demo.launch()