app.py

import gradio as gr
import joblib
import numpy as np
import pandas as pd
from huggingface_hub import hf_hub_download
from sklearn.preprocessing import StandardScaler, OneHotEncoder, LabelEncoder

# Load the trained model and scaler objects from file
REPO_ID = "Hemg/modelxxx"  # hugging face repo ID
MODEL_FILENAME = "predjob.joblib"  # model file name
SCALER_FILENAME = "scalejob.joblib"  # scaler file name

model = joblib.load(hf_hub_download(repo_id=REPO_ID, filename=MODEL_FILENAME))
scaler = joblib.load(hf_hub_download(repo_id=REPO_ID, filename=SCALER_FILENAME))

def encode_categorical_columns(df):
    # Create a copy of the DataFrame to avoid modifying the original
    df = df.copy()
    
    # Define the expected categorical columns and their order
    categorical_columns = [
        "Location", "Course", "Faculty", "College", "Source", "Event", 
        "Presenter", "Visited Parent", "Visited College for Inquiry", 
        "Attended Any Event"
    ]
    
    # Define the expected numeric columns
    numeric_columns = ["College Fee", "GPA", "Year"]
    
    # Create label encoder dictionary
    label_encoders = {}
    
    # Encode each categorical column
    for col in categorical_columns:
        label_encoders[col] = LabelEncoder()
        df[col] = label_encoders[col].fit_transform(df[col].astype(str))
    
    # Ensure numeric columns are float type
    for col in numeric_columns:
        df[col] = df[col].astype(float)
    
    # Ensure columns are in the correct order
    df = df[categorical_columns + numeric_columns]
    
    return df

def predict_performance(Location, Course, Faculty, College, Source, Event, Presenter, Visited_Parent, 
                       Visited_College_for_Inquiry, Attended_Any_Event, College_Fee, GPA, Year):
    # Create input DataFrame with consistent column names
    input_data = {
        "Location": [Location],
        "Course": [Course],
        "Faculty": [Faculty],
        "College": [College],
        "Source": [Source],
        "Event": [Event],
        "Presenter": [Presenter],
        "Visited Parent": [Visited_Parent],
        "Visited College for Inquiry": [Visited_College_for_Inquiry],
        "Attended Any Event": [Attended_Any_Event],
        "College Fee": [float(College_Fee)],
        "GPA": [float(GPA)],
        "Year": [float(Year)]
    }
    
    input_df = pd.DataFrame(input_data)
    
    # Debug print 2: Show DataFrame before encoding
    print("\nDataFrame before encoding:")
    print(input_df)
    
    # Encode categorical columns
    df = encode_categorical_columns(input_df)
    
    # Debug print 3: Show DataFrame after encoding
    print("\nDataFrame after encoding:")
    print(df)
    
    # Ensure the DataFrame columns match the order used during scaler fitting
    expected_columns = ["Location", "Course", "Faculty", "College", "Source", "Event", 
                        "Presenter", "Visited Parent", "Visited College for Inquiry", 
                        "Attended Any Event", "College Fee", "GPA", "Year"]
    df = df[expected_columns]
    
    # Scale input data using the loaded scaler
    scaled_input = scaler.transform(df)
    
    # Make the prediction
    prediction = model.predict(scaled_input)[0]
    
    # Clip the prediction to be between 0 and 1
    prediction = np.clip(prediction, 0, 1)
    
    # Debug print
    print("\nPrediction details:")
    print(f"Raw prediction: {prediction}")
    
    return f"Chance of Admission: {prediction:.1f}"

# Create Gradio interface
iface = gr.Interface(
    fn=predict_performance,
    inputs=[
        gr.Radio(["Kathmandu", "Bhaktapur", "Lalitpur", "Kritipur"], label="Location"),
        gr.Radio(["MSc IT & Applied Security", "BSc (Hons) Computing", 
                 "BSc (Hons) Computing with Artificial Intelligence", 
                 "BSc (Hons) Computer Networking & IT Security", 
                 "BSc (Hons) Multimedia Technologies", "MBA", 
                 "BA (Hons) Accounting & Finance", 
                 "BA (Hons) Business Administration"], label="Course"),
        gr.Radio(["Science", "Management", "Humanities"], label="Faculty"),
        gr.Radio(["Trinity", "CCRC", "KMC", "SOS", "ISMT", "St. Xavier's", "Everest", "Prime"], label="College"),
        gr.Radio(["Event", "Facebook", "Instagram", "Offline", "Recommendation"], label="Source"),
        gr.Radio(["New Year", "Dashain", "Orientation", "Fresher's Party", 
                 "Holi Festival", "Welcome Ceremony"], label="Event"),
        gr.Radio(["Ram", "Gita", "Manish", "Shyam", "Raj", "Hari", "Rina", "Shree"], 
                label="Presenter"),
        gr.Radio(["Yes", "No"], label="Visited Parent"),
        gr.Radio(["Trinity", "CCRC", "KMC", "SOS", "ISMT", "St. Xavier's", "Everest", "Prime"], label="College"),
        gr.Radio(["Yes", "No"], label="Visited College for Inquiry"),
        gr.Radio(["Yes", "No"], label="Attended Any Event"),
        gr.Number(label="College Fee"),
        gr.Number(label="GPA"),
        gr.Number(label="Year")
    ],
    outputs="text",
    title="Chance of Student Admission",
    description="Predict the chances of a student's admission based on various inputs."
)

# Run the app
if __name__ == "__main__":
    iface.launch(share=True)