app.py

iimport pickle
import gradio as gr

# Define the function to read the pickled model
def read_pickle(path, saved_model_name):
    with open(path + saved_model_name + '.pickle', 'rb') as to_read:
        model = pickle.load(to_read)
    return model

# Load the pickled model
path = './'  # Assuming the model file is in the current directory
model = read_pickle(path, "Automatidata_gui")

# Define the function for making predictions
def automatidata(VendorID, passenger_count, Distance, Duration, rush_hour):
    inputs = [[VendorID, passenger_count, Distance, Duration, rush_hour]]
    prediction = model.predict(inputs) 
    prediction_value = prediction[0][0]
    return f"Fare amount(approx.) = {round(prediction_value, 2)} $"

# Create the Gradio interface
automatidata_ga = gr.Interface(fn=automatidata, 
                               inputs=[
                                   gr.Number(1, 2, label="VendorID - [1, 2]"),
                                   gr.Number(0, 6, label="Passenger Count"),
                                   gr.Number(label="Distance"),
                                   gr.Number(label="Duration"),
                                   gr.Number(0, 1, label="Rush Hour")
                               ],
                               outputs="text", title="Taxi Fares Estimator", 
                               description="Predicting Taxi Fare Amount Using Machine Learning."
                               )

# Launch the interface
if __name__ == "__main__":
    automatidata_ga.launch(share=True)