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, LabelEncoder REPO_ID = "Hemg/modelxxx" MoDEL_FILENAME = "studentpredict.joblib" SCALER_FILENAME = "studentscaler.joblib" 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): label_encoder = LabelEncoder() ordinal_columns = df.select_dtypes(include=['object']).columns for col in ordinal_columns: df[col] = label_encoder.fit_transform(df[col]) nominal_columns = df.select_dtypes(include=['object']).columns.difference(ordinal_columns) df = pd.get_dummies(df, columns=nominal_columns, drop_first=True) return df def predict_performance(Location, College_Fee, College, GPA, Year, Course_Interested, Faculty, Source, Visited_College_for_Inquiry_Only, Event, Attended_Any_Events, Presenter, Visited_Parents): try: input_data = [[Location, College_Fee, College, GPA, Year, Course_Interested, Faculty, Source, Visited_College_for_Inquiry_Only, Event, Attended_Any_Events, Presenter, Visited_Parents]] feature_names = ["Location", "College Fee", "College", "GPA", "Year", "Course Interested", "Faculty", "Source", "Visited College for Inquiry Only", "Event", "Attended Any Events", "Presenter", "Visited Parents"] input_df = pd.DataFrame(input_data, columns=feature_names) df = encode_categorical_columns(input_df) df = df.reindex(columns=scaler.feature_names_in_, fill_value=0) scaled_input = scaler.transform(df) # Get probability prediction probabilities = model.predict_proba(scaled_input)[0] # Take the probability of positive class (usually index 1) admission_probability = probabilities[1] # Ensure the probability is between 0 and 1 admission_probability = np.clip(admission_probability, 0, 1) # Convert to percentage prediction_percentage = admission_probability * 100 # Create styled HTML output # html_template = """ #
#
# Admission Probability: {:.1f}% #
#
# {message} #
#
# """ # Create styled HTML output html_template = """
{message}
""" if prediction_percentage > 50: style = "font-family: Arial, sans-serif; font-size: 32px; color: #28a745; font-weight: bold;" message = "High chance of admission" elif prediction_percentage < 50: style = "font-family: Arial, sans-serif; font-size: 32px; color: #dc3545; font-weight: bold; text-transform: uppercase;" message = "Lower chance of admission" else: # exactly 50 style = "font-family: Arial, sans-serif; font-size: 32px; color: #ffc107; font-weight: bold;" message = "Moderate chance of admission" return html_template.format(prediction_percentage, style=style, message=message) except Exception as e: return f"
Error in prediction: {str(e)}
" # Update the Gradio interface iface = gr.Interface( fn=predict_performance, inputs=[ gr.Radio(["Kathmandu", "Bhaktapur", "Lalitpur", "Kritipur"], label="Location",info="What is your current location?"), gr.Slider(minimum=1000000, maximum=1700000,step=100000,label="College Fee", info="What 's the the total bachelor fee for the course you want to enroll?"), gr.Radio(["Trinity", "CCRC", "KMC", "SOS", "ISMT", "St. Xavier's", "Everest", "Prime"], label="College", info="What is the name of the last college you attended?"), gr.Slider(minimum=2, maximum=3, label="GPA", info="What is your GPA (Grade Point Average) of +2 ?"), gr.Slider(minimum=2024, maximum=2026, step=1, label="Year", info="What is your intended year of admission?"), #gr.Radio([2024, 2025, 2026], label="Year", info="What is your intended year of admission?") 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_Interested", info="Which course are you most interested in?"), gr.Radio(["Science", "Management", "Humanities"], label="Faculty", info="what is your last stream ?"), gr.Radio(["Event", "Facebook", "Instagram", "Offline", "Recommendation"], label="Source",info="How did you first hear about this college?"), gr.Radio(["Yes", "No"], label="visited_college_for_inquery_only", info="Have you visited the college you're interested in for an inquiry or consultation?"), gr.Radio(["Yes", "No"], label="attended_any_event", info="Have you attended any events organized by the college you're interested in?"), gr.Radio(["New Year", "Dashain", "Orientation", "Fresher's Party", "Holi Festival", "Welcome Ceremony"], label="attended_event_name", info="If yes, which events did you attend?" ), gr.Radio(["Ram", "Gita", "Manish", "Shyam", "Raj", "Hari", "Rina", "Shree"], label="Presenter", info="who is the counser that help you while in counseling?"), gr.Radio(["Yes", "No"], label="visited_with_parents", info="Did you visit the college with your parents?") ], outputs=gr.HTML(), # Changed to HTML output title="Student Admission Prediction", description="Predict the probability of student admission", css="body { font-family: Arial, sans-serif; }" ) if __name__ == "__main__": iface.launch(share=True)