app.py

import gradio as gr
import numpy as np
import tensorflow as tf

# Load the model
model = tf.keras.models.load_model('graduate_prediction.h5')

# Define the prediction function
def predict(curricular_units_2nd_sem_approved, curricular_units_2nd_sem_grade, curricular_units_1st_sem_approved, curricular_units_1st_sem_grade,
            curricular_units_2nd_sem_evaluations, curricular_units_1st_sem_evaluations, admission_grade, tuition_fees_up_to_date,
            previous_qualification_grade, scholarship_holder):
    # Prepare the input data (ensure it's in the correct shape)
    input_data = np.array([[curricular_units_2nd_sem_approved, curricular_units_2nd_sem_grade, curricular_units_1st_sem_approved,
                            curricular_units_1st_sem_grade, curricular_units_2nd_sem_evaluations, curricular_units_1st_sem_evaluations,
                            admission_grade, tuition_fees_up_to_date, previous_qualification_grade, scholarship_holder]])
    
    # Perform prediction
    prediction = model.predict(input_data)
    
    return prediction

# Create Gradio interface
inputs = [
    gr.Number(label="Curricular units 2nd sem (approved)"),
    gr.Number(label="Curricular units 2nd sem (grade)"),
    gr.Number(label="Curricular units 1st sem (approved)"),
    gr.Number(label="Curricular units 1st sem (grade)"),
    gr.Number(label="Curricular units 2nd sem (evaluations)"),
    gr.Number(label="Curricular units 1st sem (evaluations)"),
    gr.Number(label="Admission grade"),
    gr.Number(label="Tuition fees up to date"),
    gr.Number(label="Previous qualification (grade)"),
    gr.Number(label="Scholarship holder")
]

# Correct the output definition
output = gr.Textbox(label="Predicted Result")

gr.Interface(fn=predict, inputs=inputs, outputs=output).launch()