Purnanand/Home_Price_Prediction_Python_ML_PanelHoloviz

1import panel as pn2import pickle3import json4import numpy as np5import pandas as pd6import param7import base648import warnings9 10warnings.filterwarnings('ignore')11 12pn.extension('tabulator')13pn.extension(loading_spinner='petal', loading_color='#00aa41')14 15def image_to_base64(image_path):16    with open(image_path, "rb") as f:17        return base64.b64encode(f.read()).decode("utf-8")18 19class ImgBackground(pn.reactive.ReactiveHTML):20    object = param.Parameter()21    img_base64 = param.String()22    23    _template = """24    <div id="pn-component" style="height:100%; width:100%; position:relative;">25        <div id="background" style="background-image: url(data:image/jpeg;base64,${img_base64});26                                    background-repeat: no-repeat;27                                    background-attachment: scroll;28                                    background-position: center center;29                                    background-size: cover;30                                    filter: blur(2px);31                                    opacity: 0.8;32                                    height:100%;33                                    width:100%;34                                    position:absolute;35                                    top:0;36                                    left:0;37                                    z-index:-1;">38        </div>39        <div id="content" style="position:relative; z-index:1;">40            ${object}41        </div>42    </div>43    """44    45# Path to the local background image46background_image_path = "./background.jpg"47 48# Convert image to base6449img_base64 = image_to_base64(background_image_path)50 51# Load data and model52X = pd.read_excel("./data.xlsx")53X['total_sqft'] = X['total_sqft'].astype(int)54 55sqft_list = X['total_sqft'].unique().tolist()56bhk_list = X['bhk'].unique().tolist()57bath_list = X['bath'].unique().tolist()58 59# Define the threshold60threshold_sqft = 40061 62# Create a new list with values above the threshold63sqft_list = [x for x in sqft_list if x >= threshold_sqft]64 65minsqft = min(sqft_list)66maxsqft = max(sqft_list)67 68def load_saved_artifacts():69    global datacolumns, locations, model70    71    with open('columns.json', 'r') as f:72        datacolumns = json.load(f)['data_columns']73        locations = datacolumns[4:]74    75    with open('./banglore_home_prices_model.pickle', 'rb') as file:76        model = pickle.load(file)77 78load_saved_artifacts()79 80# Building widgets81location_select = pn.widgets.Select(name='Location', options=locations, width=300, value=locations[0], align='center')82bedroom_select = pn.widgets.Select(name='Bedrooms', options=[2,3,4], width=300, value=2, align='center')83bathroom_select = pn.widgets.Select(name='Bathrooms', options=[2,3], width=300, value=3, align='center')84sqft_slider = pn.widgets.Select(name='Square Feet', options=[2000,3000,4000], width=300, value=2000, align='center')85# sqft_slider = pn.widgets.IntSlider(name='Square Feet', start=minsqft, end=maxsqft, step=100, value=3000, width=300, align='center')86 87@pn.depends(location_select, bedroom_select, bathroom_select, sqft_slider)88def get_estimated_price(location, bhk, bath, sqft):89    try:90        loc_index = datacolumns.index(location.lower())91    except:92        loc_index = -193 94    x = np.zeros(len(datacolumns))95    x[0] = sqft96    x[1] = bath97    x[2] = bhk98    if loc_index >= 0:99        x[loc_index] = 1100    101    output = round(model.predict([x])[0], 2)102    103    return pn.indicators.Number(104        name="Estimated House Price",105        value=round(output/100,2),106        format='{value} Crore Rupees',107        title_size='24pt',108        font_size='36pt',109        styles={110            'background-color': 'rgba(95, 158, 160, 0.7)',111            'border': '',112            'color': 'white',113            'padding': '10px 20px',114            'text-align': 'center',115            'text-decoration': 'none',116            'font-family': 'tahoma',117            'margin': '20px auto',118            'cursor': 'default',119            'border-radius': '10px',120            'box-shadow': '0 4px 6px rgba(0, 0, 0, 0.1)',121            'width': '300px'122        }123    )124 125component1 = pn.Column(126    pn.Row(pn.Spacer(width=300),pn.pane.Markdown("# Housing Price Prediction Model", styles={127                'background-color': '#F0FFFF',128                'border': '',129                'color': 'black',130                'padding': '5px 5px',131                'text-align': 'center',132                'text-decoration': 'none',133                'font-family': 'tahoma',134                'margin': '10px auto',135                'cursor': 'default',136                'font-size': '20px',137                'font-weight': 'bold',138            })),139    pn.Row(140        pn.Column(141            pn.pane.Markdown("## Select a location", styles={142                'background-color': '#F0FFFF',143                'border': '',144                'color': 'black',145                'padding': '5px 5px',146                'text-align': 'center',147                'text-decoration': 'none',148                'font-family': 'tahoma',149                'margin': '10px auto',150                'cursor': 'default',151                'font-size': '10px',152                'font-weight': 'bold',153            }),154            pn.Row(location_select, width=300, align='center'),155            pn.Spacer(height=30),156            pn.pane.Markdown("## Select number of bedrooms", styles={157                'background-color': '#F0FFFF',158                'border': '',159                'color': 'black',160                'padding': '5px 5px',161                'text-align': 'center',162                'text-decoration': 'none',163                'font-family': 'tahoma',164                'margin': '10px auto',165                'cursor': 'default',166                'font-size': '10px',167                'font-weight': 'bold',168            }),169            pn.Row(bedroom_select, width=300, align='center'),170            pn.Spacer(height=30),171            pn.pane.Markdown("## Select number of bathrooms", styles={172                'background-color': '#F0FFFF',173                'border': '',174                'color': 'black',175                'padding': '5px 5px',176                'text-align': 'center',177                'text-decoration': 'none',178                'font-family': 'tahoma',179                'margin': '10px auto',180                'cursor': 'default',181                'font-size': '10px',182                'font-weight': 'bold',183            }),184            pn.Row(bathroom_select, width=300, align='center'),185            pn.Spacer(height=30),186            pn.pane.Markdown("## Select square feet using the slider", styles={187                'background-color': '#F0FFFF',188                'border': '',189                'color': 'black',190                'padding': '5px 5px',191                'text-align': 'center',192                'text-decoration': 'none',193                'font-family': 'tahoma',194                'margin': '10px auto',195                'cursor': 'default',196                'font-size': '10px',197                'font-weight': 'bold',198            }),199            pn.Row(sqft_slider, width=300, align='center'),200            align='center',201            width=600202        ),203        pn.Column(204            pn.pane.Markdown("## Predicted Price", styles={205                'background-color': '#F0FFFF',206                'border': '',207                'color': 'black',208                'padding': '5px 5px',209                'text-align': 'center',210                'text-decoration': 'none',211                'font-family': 'tahoma',212                'margin': '10px auto',213                'cursor': 'default',214                'font-size': '10px',215                'font-weight': 'bold',216            }),217            pn.Row(get_estimated_price, width=300, align='center'),218            align='center',219            width=600220        )221    ), 222    name="Housing Price Prediction"223)224 225svg_background = ImgBackground(object=component1, height=800, img_base64=img_base64)226 227svg_background.servable()228 229