from warnings import filterwarnings
filterwarnings('ignore')
import os
import uuid
import joblib
import json
import gradio as gr
import pandas as pd
from huggingface_hub import CommitScheduler
from pathlib import Path

# print('*** Running train.py ***')
# import subprocess

# # Run the training script
# subprocess.run(["python", "train.py"])
# print('*** done! ***')

# Configure the logging functionality
log_file = Path("logs/") / f"data_{uuid.uuid4()}.json"
log_folder = log_file.parent

repo_id = "eric-green-insurance-charge-predictor-logs"

# Create a commit scheduler
scheduler = CommitScheduler(
    repo_id=repo_id,
    repo_type="dataset",
    folder_path=log_folder,
    path_in_repo="data",
    every=2
)

# Load the saved model
insurance_charge_predictor = joblib.load('model.joblib')

# Define the input features

#numeric_features = ['age', 'bmi', 'children']
#categorical_features = ['sex', 'smoker', 'region']

age_input = gr.Number(label="Age")
bmi_input = gr.Number(label="BMI")
children_input = gr.Number(label="Children")

# sex: ['female' 'male']
# smoker: ['yes' 'no']
# region: ['southwest' 'southeast' 'northwest' 'northeast']

sex_input = gr.Dropdown(['female','male'],label='Sex')
smoker_input = gr.Dropdown(['yes','no'],label='Smoker')
region_input = gr.Dropdown(['southwest', 'southeast', 'northwest', 'northeast'],label='Region')

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

# Define the predict function which will take features, convert to dataframe and make predictions using the saved model
# the functions runs when 'Submit' is clicked or when a API request is made

def predict_insurance_charges(age, bmi, children, sex, smoker, region):
    
    #Index(['age', 'sex', 'bmi', 'children', 'smoker', 'region'], dtype='object')

    sample = {
        'age': age,
        'sex': sex,
        'bmi': bmi,
        'children': children,
        'smoker': smoker,
        'region': region
    }

    data_point = pd.DataFrame([sample])
    
    prediction = insurance_charge_predictor.predict(data_point).tolist()

    with scheduler.lock:
        with log_file.open("a") as f:
            f.write(json.dumps(
                {
                    'age': age,
                    'sex': sex,
                    'bmi': bmi,
                    'children': children,
                    'smoker': smoker,
                    'region': region,
                    'prediction': prediction[0]
                }
            ))
            f.write("\n")

    return prediction[0]

gr_interface = gr.Interface(
    fn=predict_insurance_charges,
    inputs=[age_input,
            bmi_input,
            children_input,
            sex_input,
            smoker_input,
            region_input],
    outputs=model_output,
    title="HealthyLife Insurance Charge Prediction",
    description="This API allows you to predict insurance charges based on the input features.",
    allow_flagging="auto",
    concurrency_limit=8
)

gr_interface.queue()
gr_interface.launch(share=False)