app.py

# +++
import os
import uuid
import joblib
import json

# IMPORTANT: I already installed the package "gradio" in my current Virtual Environment (VEnvDSDIL_gpu_Py3.12) as:  pip install -q gradio_client
#            Do NOT install "gradio_client" package again in Anaconda otherwise it will mess up the package.
import gradio as gr
import pandas as pd

# must install the package "huggingface_hub" first in the current python Virtual Environment, with pip, not with conda, as follows
# pip install huggingface_hub
# i.e., in the command line interface within the activated Virtual Environment:
#  (VEnvDSDIL_gpu_Py3.12) epalvarez@DSDILmStation01:~ $ pip install huggingface_hub
from huggingface_hub import CommitScheduler
from pathlib import Path

# path = Path.cwd()

log_file = Path("logs/") / f"data_{uuid.uuid4()}.json"
log_folder = log_file.parent

hf_token = os.environ.get('HF_TOKEN')
print(hf_token)

# Scheduler will log every 2 API calls:
scheduler = CommitScheduler(
    repo_id="term-deposit-logs",
    repo_type="dataset",
    folder_path=log_folder,
    path_in_repo="data",
    every=2
)

term_deposit_predictor = joblib.load('model_bt.joblib')

age_input = gr.Number(label="Age")
duration_input = gr.Number(label='Duration(Sec)')
cc_contact_freq_input = gr.Number(label='CC Contact Freq')
days_since_pc_input = gr.Number(label='Days Since PC')
pc_contact_freq_input = gr.Number(label='PC Contact Freq')
job_input = gr.Dropdown(['admin.', 'blue-collar', 'technician', 'services', 'management',
        'retired', 'entrepreneur', 'self-employed', 'housemaid', 'unemployed',
        'student', 'unknown'], label="Job")
marital_status_input = gr.Dropdown(['married', 'single', 'divorced', 'unknown'], label='Marital Status')
education_input = gr.Dropdown(['experience', 'university degree', 'high school', 'professional.course',
        'Others', 'illiterate'], label='Education')
defaulter_input = gr.Dropdown(['no', 'unknown', 'yes'], label='Defaulter')
home_loan_input = gr.Dropdown(['yes', 'no', 'unknown'], label='Home Loan')
personal_loan_input = gr.Dropdown(['yes', 'no', 'unknown'], label='Personal Loan')
communication_type_input = gr.Dropdown(['cellular', 'telephone'], label='Communication Type')
last_contacted_input = gr.Dropdown(['mar', 'apr', 'may', 'jun', 'jul', 'aug', 'sep', 'oct', 'nov', 'dec'], label='Last Contacted')
day_of_week_input = gr.Dropdown(['mon', 'tue', 'wed', 'thu', 'fri'], label='Day of Week')
pc_outcome_input = gr.Dropdown(['nonexistent', 'failure', 'success'], label='PC Outcome')


model_output = gr.Label(label="Subscribed")

# -------------------------------------------------------------------------------------------------------------------------------------------------------------
def predict_term_deposit(age, duration, cc_contact_freq, days_since_pc, pc_contact_freq, job, marital_status, education, 
                            defaulter, home_loan, personal_loan, communication_type, last_contacted, 
                            day_of_week, pc_outcome):
    sample = {
        'Age': age,
        'Duration(Sec)': duration,
        'CC Contact Freq': cc_contact_freq,
        'Days Since PC': days_since_pc,
        'PC Contact Freq': pc_contact_freq,
        'Job': job,
        'Marital Status': marital_status,
        'Education': education,
        'Defaulter': defaulter,
        'Home Loan': home_loan,
        'Personal Loan': personal_loan,
        'Communication Type': communication_type,
        'Last Contacted': last_contacted,
        'Day of Week': day_of_week,
        'PC Outcome': pc_outcome,
    }
    data_point = pd.DataFrame([sample])
    prediction = term_deposit_predictor.predict(data_point).tolist()

    # Push prediction to a dataset repo for logging
    # Each time we get a prediction we will determine if we should log it to a hugging_face dataset according to the schedule definition outside this function
    with scheduler.lock:
        with log_file.open("a") as f:
            f.write(json.dumps(
                {
                    'Age': age,
                    'Duration(Sec)': duration,
                    'CC Contact Freq': cc_contact_freq,
                    'Days Since PC': days_since_pc,
                    'PC Contact Freq': pc_contact_freq,
                    'Job': job,
                    'Marital Status': marital_status,
                    'Education': education,
                    'Defaulter': defaulter,
                    'Home Loan': home_loan,
                    'Personal Loan': personal_loan,
                    'Communication Type': communication_type,
                    'Last Contacted': last_contacted,
                    'Day of Week': day_of_week,
                    'PC Outcome': pc_outcome,
                    'prediction': prediction[0]
                }
            ))
            f.write("\n")
            
    return prediction[0]
# -------------------------------------------------------------------------------------------------------------------------------------------------------------

demo = gr.Interface(
    fn=predict_term_deposit,
    inputs=[age_input,
            duration_input,
            cc_contact_freq_input,
            days_since_pc_input,
            pc_contact_freq_input,
            job_input,
            marital_status_input,
            education_input,
            defaulter_input,
            home_loan_input,
            personal_loan_input,
            communication_type_input,
            last_contacted_input,
            day_of_week_input,
            pc_outcome_input],
    outputs=model_output,
    title="Term Deposit Prediction",
    description="This API allows you to predict the person who are going to likely subscribe to the term deposit",
    allow_flagging="auto",    # automatically push to the HuggingFace Dataset
    concurrency_limit=8
)

demo.queue()
demo.launch(share=False)    # To create a public link, set "share=True" in launch() ....  but if I execute this app.py locally, then I have to have my computer on for the public users to access the browser interface