import gradio as gr
import pandas as pd
import pickle


# Define params names
PARAMS_NAME = [
            "ph",
            "Hardness",
            "Solids",
            "Chloramines",
            "Sulfate",
            "Conductivity",
            "Organic_carbon",
            "Trihalomethanes",
            "Turbidity"
]

           
# Load model
with open("model/model1.pkl", "rb") as f:
    model = pickle.load(f)




def predict(*args):
    answer_dict = {}

    for i in range(len(PARAMS_NAME)):
        answer_dict[PARAMS_NAME[i]] = [args[i]]

    # Crear dataframe
    single_instance = pd.DataFrame.from_dict(answer_dict)

 
    
    
    prediction = model.predict(single_instance)

    # Como sabemos el model nos devuelve los tipos de fraude 1, 2 y 3 en el response. Podemos devolver un response estilo semáforo.
    
    # Cast numpy.int64 to just a int
    Potability = int(prediction[0])


    # Adaptación respuesta
    response = Potability
    if Potability == 1:
        response = "Just Water, drink it 🥤"
    if Potability == 0:
        response = "Drink and you will DIE! ☠️ "


    return response


with gr.Blocks() as demo:
    gr.Markdown(
        """
        #  🏞️ Water Quality 
        """
    )

    with gr.Row():
        with gr.Column():

            gr.Markdown(
                """
                ## 🌊 Insert your water data here please 💦
                """
            )
            
            ph = gr.Slider(
                label='Ph',
                minimum=0.2274990502021987,
                maximum=13.999999999999998,
                step=0.1,
                randomize=True
            )

            Hardness = gr.Slider(
                label='Hardness',
                minimum=73.4922336890611,
                maximum=317.33812405558257,
                step=0.1,
                randomize=True
            )

            Solids = gr.Slider(
                label='Solids',
                minimum=320.942611274359,
                maximum=56488.67241273919,
                step=0.1,
                randomize=True
            )

            Chloramines = gr.Slider(
                label='Chloramines',
                minimum=1.3908709048851806,
                maximum=13.127000000000002,
                step=0.1,
                randomize=True
            )

            Sulfate = gr.Slider(
                label='Sulfate',
                minimum=129.00000000000003,
                maximum=481.0306423059972,
                step=0.1,
                randomize=True
            )

            Conductivity = gr.Slider(
                label='Conductivity',
                minimum=201.6197367551575,
                maximum=753.3426195583046,
                step=0.1,
                randomize=True
            )

            Organic_carbon = gr.Slider(
                label='Organic_Carbon',
                minimum=2.1999999999999886,
                maximum=27.00670661116601,
                step=0.1,
                randomize=True
            )

            Trihalomethanes = gr.Slider(
                label='Trihalomethanes',
                minimum=8.577012932983806,
                maximum=124.0,
                step=0.1,
                randomize=True
            )

            Turbidity = gr.Slider(
                label='Turbidity',
                minimum=1.45,
                maximum=6.494748555990993,
                step=0.1,
                randomize=True
            )


        with gr.Column():

            gr.Markdown(
                """
                ## Let's see if you can drink it 💧
                """
            )

            label = gr.Label(label="Kind of Water")
            predict_btn = gr.Button(value="Test")
            predict_btn.click(
                predict,
                inputs=[
                    ph,
                    Hardness,
                    Solids,
                    Chloramines,
                    Sulfate,
                    Conductivity,
                    Organic_carbon,
                    Trihalomethanes,
                    Turbidity,
                ],
                outputs=[label],
                api_name="prediccion"
            )
    gr.Markdown(
        """
        <p style='text-align: center'>
            <a href='https://www.escueladedatosvivos.ai/cursos/bootcamp-de-data-science' 
                target='_blank'>Proyecto demo creado en el bootcamp de EDVAI 🤗
            </a>
        </p>
        <p style='text-align: center'>
            <a href='https://www.kaggle.com/datasets/adityakadiwal/water-potability' 
                target='_blank'>Data From Water Quality in Kaggle
            </a>
        </p>
        """
    )

demo.launch()