app.py

import pickle
import gradio as gr

# Load the saved models
def load_models():
    try:
        diabetes_model = pickle.load(open('Saved Models/diabetes_model.sav', 'rb'))
        heart_disease_model = pickle.load(open('Saved Models/heart_disease_model.sav', 'rb'))
        parkinsons_model = pickle.load(open('Saved Models/parkinsons_model.sav', 'rb'))
        return diabetes_model, heart_disease_model, parkinsons_model
    except Exception as e:
        return None, None, None

# Function to make prediction
def make_prediction(model, input_data):
    try:
        prediction = model.predict([input_data])
        return prediction[0]
    except Exception as e:
        return None

# Functions for each prediction
def predict_diabetes(Pregnancies, Glucose, BloodPressure, SkinThickness, Insulin, BMI, DiabetesPedigreeFunction, Age):
    input_data = [Pregnancies, Glucose, BloodPressure, SkinThickness, Insulin, BMI, DiabetesPedigreeFunction, Age]
    
    result = make_prediction(diabetes_model, input_data)
    return "Diabetic" if result == 1 else "Not Diabetic"

def predict_heart_disease(age, sex, cp, trestbps, chol, fbs, restecg, thalach, exang, oldpeak, slope, ca, thal):
    input_data = [age, sex, cp, trestbps, chol, fbs, restecg, thalach, exang, oldpeak, slope, ca, thal]
    if input_data[1] == 'male':
        input_data[1] = 1
    elif input_data[1] == 'female':
        input_data[1] = 0
    else:
        input_data[1] = None
    
    result = make_prediction(heart_disease_model, input_data)
    return "Has Heart Disease" if result == 1 else "No Heart Disease"

def predict_parkinsons(fo, fhi, flo, Jitter_percent, Jitter_Abs, RAP, PPQ, DDP, Shimmer, Shimmer_dB, APQ3, APQ5, APQ, DDA, NHR, HNR, RPDE, DFA, spread1, spread2, D2, PPE):
    input_data = [fo, fhi, flo, Jitter_percent, Jitter_Abs, RAP, PPQ, DDP, Shimmer, Shimmer_dB, APQ3, APQ5, APQ, DDA, NHR, HNR, RPDE, DFA, spread1, spread2, D2, PPE]
    result = make_prediction(parkinsons_model, input_data)
    return "Has Parkinson's Disease" if result == 1 else "No Parkinson's Disease"

# Load models
diabetes_model, heart_disease_model, parkinsons_model = load_models()

if not diabetes_model or not heart_disease_model or not parkinsons_model:
    print("Error loading models. Please check the model paths.")

# Define Gradio interfaces for each disease prediction
diabetes_interface = gr.Interface(
    fn=predict_diabetes,
    inputs=[
        gr.Text(label="Number of Pregnancies"),
        gr.Number(label="Glucose Level"),
        gr.Number(label="Blood Pressure"),
        gr.Number(label="Skin Thickness"),
        gr.Number(label="Insulin Level"),
        gr.Number(label="BMI"),
        gr.Number(label="Diabetes Pedigree Function"),
        gr.Number(label="Age"),
    ],
    outputs="text",
    title="Diabetes Prediction",
    description="Predict if a person has diabetes based on medical test results."
)

heart_disease_interface = gr.Interface(
    fn=predict_heart_disease,
    inputs=[
        gr.Number(label="Age"),
        gr.Text(label="Sex (Male, Female)"),
        gr.Number(label="Chest Pain types (0-3)"),
        gr.Number(label="Resting Blood Pressure"),
        gr.Number(label="Serum Cholestoral in mg/dl"),
        gr.Number(label="Fasting Blood Sugar > 120 mg/dl (1=True, 0=False)"),
        gr.Number(label="Resting Electrocardiographic results (0-2)"),
        gr.Number(label="Maximum Heart Rate achieved"),
        gr.Number(label="Exercise Induced Angina (1=Yes, 0=No)"),
        gr.Number(label="ST depression induced by exercise"),
        gr.Number(label="Slope of the peak exercise ST segment (0-2)"),
        gr.Number(label="Major vessels colored by flourosopy (0-3)"),
        gr.Number(label="Thalassemia (1=normal; 2=fixed defect; 3=reversible defect)"),
    ],
    outputs="text",
    title="Heart Disease Prediction",
    description="Predict if a person has heart disease based on medical test results."
)

parkinsons_interface = gr.Interface(
    fn=predict_parkinsons,
    inputs=[
        gr.Number(label="MDVP:Fo(Hz)"),
        gr.Number(label="MDVP:Fhi(Hz)"),
        gr.Number(label="MDVP:Flo(Hz)"),
        gr.Number(label="MDVP:Jitter(%)"),
        gr.Number(label="MDVP:Jitter(Abs)"),
        gr.Number(label="MDVP:RAP"),
        gr.Number(label="MDVP:PPQ"),
        gr.Number(label="Jitter:DDP"),
        gr.Number(label="MDVP:Shimmer"),
        gr.Number(label="MDVP:Shimmer(dB)"),
        gr.Number(label="Shimmer:APQ3"),
        gr.Number(label="Shimmer:APQ5"),
        gr.Number(label="MDVP:APQ"),
        gr.Number(label="Shimmer:DDA"),
        gr.Number(label="NHR"),
        gr.Number(label="HNR"),
        gr.Number(label="RPDE"),
        gr.Number(label="DFA"),
        gr.Number(label="spread1"),
        gr.Number(label="spread2"),
        gr.Number(label="D2"),
        gr.Number(label="PPE"),
    ],
    outputs="text",
    title="Parkinson's Disease Prediction",
    description="Predict if a person has Parkinson's disease based on voice measurements."
)

# Combine interfaces into a tabbed layout
gr.TabbedInterface(
    [diabetes_interface, heart_disease_interface, parkinsons_interface],
    ["Diabetes Prediction", "Heart Disease Prediction", "Parkinson's Disease Prediction"]
).launch()