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

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from warnings import filterwarnings
filterwarnings('ignore')


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

scheduler = CommitScheduler(
    repo_id="machine-failure-logs",
    repo_type="dataset",
    folder_path=log_folder,
    path_in_repo="data",
    every=2
)

health_status_predictor = joblib.load('model.joblib')

Latitude = gr.Number(label='Latitude')
Longitude = gr.Number(label='Longitude')
DBH = gr.Number(label='DBH')
Tree_Height = gr.Number(label='Tree_Height')
Crown_Width_North_South = gr.Number(label='Crown_Width_North_South')
Crown_Width_East_West = gr.Number(label='Crown_Width_East_West')
Slope = gr.Number(label='Slope')
Elevation = gr.Number(label='Elevation')
Temperature = gr.Number(label='Temperature')
Humidity = gr.Number(label='Humidity')
Soil_TN = gr.Number(label='Soil_TN')
Soil_TP = gr.Number(label='Soil_TP')
Soil_AP = gr.Number(label='Soil_AP')
Soil_AN = gr.Number(label='Soil_AN')
Menhinick_Index = gr.Number(label='Menhinick_Index')
Gleason_Index = gr.Number(label='Gleason_Index')
Fire_Risk_Index = gr.Number(label='Fire_Risk_Index')

model_output = gr.Label(label="Health Status")

def predict_health_status(Latitude, Longitude, DBH, Tree_Height, Crown_Width_North_South, Crown_Width_East_West, Slope, Elevation, Temperature, Humidity, Soil_TN, Soil_TP, Soil_AP, Soil_AN, Menhinick_Index, Gleason_Index, Fire_Risk_Index):
    sample = {
        'Latitude': Latitude,
        'Longitude': Longitude,
        'DBH': DBH,
        'Tree_Height': Tree_Height,
        'Crown_Width_North_South': Crown_Width_North_South,
        'Crown_Width_East_West': Crown_Width_East_West,
        'Slope': Slope,
        'Elevation': Elevation,
        'Temperature': Temperature,
        'Humidity': Humidity,
        'Soil_TN': Soil_TN,
        'Soil_TP': Soil_TP,
        'Soil_AP': Soil_AP,
        'Soil_AN': Soil_AN,
        'Menhinick_Index': Menhinick_Index,
        'Gleason_Index': Gleason_Index.
        'Fire_Risk_Index': Fire_Risk_Index
        
    }
    data_point = pd.DataFrame([sample])
    prediction = health_status_predictor.predict(data_point).tolist()

    with scheduler.lock:
        with log_file.open("a") as f:
            f.write(json.dumps(
                {
                    'Latitude': Latitude,
                    'Longitude': Longitude,
                    'DBH': DBH,
                    'Tree_Height': Tree_Height,
                    'Crown_Width_North_South': Crown_Width_North_South,
                    'Crown_Width_East_West': Crown_Width_East_West,
                    'Slope': Slope,
                    'Elevation': Elevation,
                    'Temperature': Temperature,
                    'Humidity': Humidity,
                    'Soil_TN': Soil_TN,
                    'Soil_TP': Soil_TP,
                    'Soil_AP': Soil_AP,
                    'Soil_AN': Soil_AN,
                    'Menhinick_Index': Menhinick_Index,
                    'Gleason_Index': Gleason_Index,
                    'Fire_Risk_Index': Fire_Risk_Index,
                    'prediction': prediction[0]
                }
            ))
            f.write("\n")
            
    return prediction[0]

demo = gr.Interface(
    fn=predict_health_status,
    inputs=[Latitude, Longitude, DBH, Tree_Height, Crown_Width_North_South, Crown_Width_East_West, Slope, Elevation, Temperature, Humidity, Soil_TN, Soil_TP, Soil_AP, Soil_AN, Menhinick_Index, Gleason_Index, Fire_Risk_Index],
    outputs=model_output,
    title="Health Status Predictor",
    description="This API allows you to predict the health status of a Tree",
    allow_flagging="auto",
    concurrency_limit=8
)


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