app.py

import gradio as gr
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import os
from sklearn.ensemble import RandomForestRegressor
from PIL import Image

# --------------------------------------------------
# LOGO PATH
# --------------------------------------------------
LOGO_PATH = r"C:\Users\SASTRA\Desktop\sastra_logo.jpg"

logo_img = Image.open(LOGO_PATH) if os.path.exists(LOGO_PATH) else None

CSV_FILE = "student_analysis_database.csv"

# --------------------------------------------------
# MODEL TRAINING
# --------------------------------------------------
np.random.seed(42)

X_train = np.random.uniform(1,10,(400,15))
y_train = (
    0.15*X_train[:,1] +
    0.15*X_train[:,2] +
    0.18*X_train[:,3] +
    0.14*X_train[:,4] +
    0.10*X_train[:,5] +
    0.10*X_train[:,6] +
    0.18*X_train[:,7]
) * 8

model = RandomForestRegressor(n_estimators=150, random_state=42)
model.fit(X_train,y_train)

# --------------------------------------------------
# IQ CALCULATION
# --------------------------------------------------
def calculate_iq(reasoning, aptitude, problem_solving,
                 verbal, communication, understanding):

    return round((
        0.25*reasoning +
        0.20*aptitude +
        0.20*problem_solving +
        0.15*verbal +
        0.10*communication +
        0.10*understanding
    ) * 10, 2)

# --------------------------------------------------
# MAIN FUNCTION
# --------------------------------------------------
def institutional_ai(
    name, regno, vision, mission, arrears,
    sg1, sg2, sg3, sg4, sg5, sg6,
    cgpa,
    understanding, coding, problem_solving,
    presentation, aptitude, reasoning,
    verbal, communication, team_building,
    group_discussion, study_time):

    iq = calculate_iq(reasoning, aptitude,
                      problem_solving, verbal,
                      communication, understanding)

    features = np.array([[
        arrears, sg1, sg2, sg3, sg4, sg5,
        sg6, cgpa, coding, problem_solving,
        aptitude, communication, iq,
        study_time, presentation
    ]])

    performance = float(model.predict(features)[0])

    weak=[]
    if coding<5: weak.append("Coding")
    if aptitude<5: weak.append("Aptitude")
    if communication<5: weak.append("Communication")

    concentration=", ".join(weak) if weak else "Balanced Skill Profile"

    placement = (
        "High Probability → Product Companies"
        if performance>=80 and cgpa>=8
        else "Moderate Probability → Service Companies"
        if performance>=65
        else "Needs Skill Improvement"
    )

    advice="Improve weak areas and maintain academic consistency."

    # ---------- Graphs ----------
    fig1=plt.figure()
    plt.bar(["Performance"],[performance])
    plt.ylim(0,100)

    fig2=plt.figure()
    plt.plot(range(1,7),[sg1,sg2,sg3,sg4,sg5,sg6],marker='o')

    labels=['Understanding','Coding','ProblemSolving',
            'Aptitude','Communication','Presentation']
    skills=[understanding,coding,problem_solving,
            aptitude,communication,presentation]

    angles=np.linspace(0,2*np.pi,len(labels),endpoint=False)
    skills=np.concatenate((skills,[skills[0]]))
    angles=np.concatenate((angles,[angles[0]]))

    fig3=plt.figure()
    ax=fig3.add_subplot(111,polar=True)
    ax.plot(angles,skills)
    ax.fill(angles,skills,alpha=0.2)

    report=f"Student: {name}\nPerformance: {round(performance,2)}\nIQ: {iq}"

    return performance, iq, concentration, placement, advice, report, fig1, fig2, fig3


# --------------------------------------------------
# UI USING BLOCKS ONLY (KEY FIX)
# --------------------------------------------------
with gr.Blocks() as demo:

    # LOGO CENTERED
    if logo_img is not None:
        gr.Image(value=logo_img, show_label=False, height=160)

    gr.Markdown(
        """
        <div style='text-align:center'>
        <h2>Srinivasa Ramanujan Centre,<br>
        SASTRA Deemed to be University, Kumbakonam</h2>
        <h3>Student Performance Analysis</h3>
        </div>
        """
    )

    # INPUTS
    name = gr.Textbox(label="Student Name")
    regno = gr.Textbox(label="Register Number")
    vision = gr.Textbox(label="Vision")
    mission = gr.Textbox(label="Mission")
    arrears = gr.Slider(0,10,label="Arrears")

    sg1 = gr.Slider(0,10,label="SGPA 1")
    sg2 = gr.Slider(0,10,label="SGPA 2")
    sg3 = gr.Slider(0,10,label="SGPA 3")
    sg4 = gr.Slider(0,10,label="SGPA 4")
    sg5 = gr.Slider(0,10,label="SGPA 5")
    sg6 = gr.Slider(0,10,label="SGPA 6")

    cgpa = gr.Slider(0,10,label="CGPA")

    understanding = gr.Slider(1,10,label="Understanding")
    coding = gr.Slider(1,10,label="Coding")
    problem_solving = gr.Slider(1,10,label="Problem Solving")
    presentation = gr.Slider(1,10,label="Presentation")
    aptitude = gr.Slider(1,10,label="Aptitude")
    reasoning = gr.Slider(1,10,label="Reasoning")
    verbal = gr.Slider(1,10,label="Verbal")
    communication = gr.Slider(1,10,label="Communication")
    team_building = gr.Slider(1,10,label="Team Building")
    group_discussion = gr.Slider(1,10,label="Group Discussion")
    study_time = gr.Slider(1,10,label="Daily Study Time")

    submit = gr.Button("Analyze Student")

    # OUTPUTS
    out1 = gr.Number(label="Predicted Performance")
    out2 = gr.Number(label="Estimated IQ")
    out3 = gr.Textbox(label="Concentration Areas")
    out4 = gr.Textbox(label="Placement")
    out5 = gr.Textbox(label="Advice")
    out6 = gr.Textbox(label="Report")
    plot1 = gr.Plot()
    plot2 = gr.Plot()
    plot3 = gr.Plot()

    submit.click(
        institutional_ai,
        inputs=[name,regno,vision,mission,arrears,
                sg1,sg2,sg3,sg4,sg5,sg6,cgpa,
                understanding,coding,problem_solving,
                presentation,aptitude,reasoning,
                verbal,communication,team_building,
                group_discussion,study_time],
        outputs=[out1,out2,out3,out4,out5,out6,plot1,plot2,plot3]
    )

demo.launch(inbrowser=True)