Spaces:
Runtime error
Runtime error
| import streamlit as st | |
| import pandas as pd | |
| import numpy as np | |
| from sklearn.linear_model import LinearRegression | |
| from sklearn.model_selection import train_test_split | |
| from sklearn.metrics import mean_absolute_error, r2_score | |
| import plotly.express as px | |
| import plotly.graph_objects as go | |
| # Configuration de la page | |
| st.set_page_config( | |
| page_title="IA Prédilo - Immobilier", | |
| page_icon="🏠", | |
| layout="wide" | |
| ) | |
| # Style CSS pour une interface premium | |
| st.markdown(""" | |
| <style> | |
| .prediction-box { | |
| background-color: #ffffff; | |
| padding: 25px; | |
| border-radius: 15px; | |
| border-left: 10px solid #007bff; | |
| box-shadow: 0 4px 15px rgba(0,0,0,0.1); | |
| margin: 20px 0; | |
| } | |
| </style> | |
| """, unsafe_allow_html=True) | |
| # ========================================================== | |
| # 1. LE JEU DE DONNÉES (Dataset - 30 exemples exacts) | |
| # ========================================================== | |
| def load_data(): | |
| donnees = { | |
| 'm2': [30, 50, 70, 90, 110, 130, 30, 50, 70, 50, 45, 85, 120, 60, 95, 55, 75, 100, 65, 80, 40, 90, 115, 140, 35, 60, 95, 105, 125, 150], | |
| 'dist': [1, 2, 3, 5, 8, 10, 15, 20, 25, 2, 4, 6, 12, 5, 8, 3, 4, 7, 2, 5, 1, 6, 9, 11, 18, 3, 7, 8, 10, 13], | |
| 'neuf': [0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 1], | |
| 'pieces': [1, 2, 3, 4, 4, 5, 1, 2, 3, 2, 2, 3, 4, 3, 4, 2, 3, 4, 3, 3, 1, 4, 5, 6, 1, 2, 4, 4, 5, 6], | |
| 'parking': [0, 1, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1], | |
| 'etage': [0, 2, 1, 3, 0, 1, 0, 1, 0, 4, 2, 3, 0, 2, 5, 3, 1, 2, 5, 4, 0, 3, 1, 0, 2, 4, 6, 2, 1, 3], | |
| 'balcon': [0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1], | |
| 'jardin': [0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1], | |
| 'ascenseur':[0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 1, 1], | |
| 'dpe': [4, 3, 4, 3, 5, 4, 6, 5, 6, 2, 4, 3, 1, 4, 3, 2, 4, 2, 1, 3, 5, 2, 3, 4, 6, 2, 3, 1, 3, 1], | |
| 'annee': [1985, 2000, 1995, 2005, 1980, 1990, 1975, 1988, 1970, 2020, 1998, 2008, 2022, 2002, 2010, 2018, 1992, 2015, 2021, 2005, 1982, 2012, 2003, 1995, 1978, 2019, 2011, 2023, 2007, 2024], | |
| 'etat': [2, 2, 2, 2, 1, 2, 1, 1, 1, 3, 2, 2, 3, 2, 2, 3, 2, 3, 3, 2, 1, 3, 2, 2, 1, 3, 2, 3, 2, 3], | |
| 'prix': [180000, 275000, 362000, 418000, 485000, 515000, 115000, 182000, 238000, 315000, 220000, 380000, 540000, 290000, 430000, 298000, 355000, 475000, 325000, 395000, 195000, 445000, 520000, 580000, 145000, 310000, 460000, 495000, 535000, 625000] | |
| } | |
| return pd.DataFrame(donnees) | |
| df = load_data() | |
| # ========================================================== | |
| # 2. ENTRAÎNEMENT DE L'IA (Train / Test Split) | |
| # ========================================================== | |
| features = ['m2', 'dist', 'neuf', 'pieces', 'parking', 'etage', 'balcon', 'jardin', 'ascenseur', 'dpe', 'annee', 'etat'] | |
| X = df[features] | |
| y = df['prix'] | |
| X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) | |
| model = LinearRegression() | |
| model.fit(X_train, y_train) | |
| y_pred_test = model.predict(X_test) | |
| mae_test = mean_absolute_error(y_test, y_pred_test) | |
| score_r2 = r2_score(y_test, y_pred_test) | |
| # ========================================================== | |
| # 3. INTERFACE (Sidebar & Menu) | |
| # ========================================================== | |
| st.sidebar.header("🔧 Paramètres du bien") | |
| surface = st.sidebar.slider("Surface (m2)", 20, 300, 75) | |
| pieces = st.sidebar.number_input("Nombre de pièces", 1, 10, 3) | |
| distance = st.sidebar.slider("Distance centre (km)", 0, 50, 5) | |
| neuf_o_n = st.sidebar.selectbox("Est-ce neuf ?", ["NON", "OUI"]) | |
| annee = st.sidebar.number_input("Année de construction", 1900, 2025, 2010) | |
| etat = st.sidebar.select_slider("État général", options=["A rénover", "Bon", "Excellent"], value="Bon") | |
| dpe_label = st.sidebar.select_slider("Classe DPE", options=["A", "B", "C", "D", "E", "F", "G"], value="C") | |
| st.sidebar.subheader("Équipements") | |
| parking = st.sidebar.checkbox("Parking / Garage", value=True) | |
| balcon = st.sidebar.checkbox("Balcon / Terrasse") | |
| jardin = st.sidebar.checkbox("Jardin") | |
| ascenseur = st.sidebar.checkbox("Ascenseur") | |
| prix_vendeur = st.sidebar.number_input("Prix du vendeur (€)", 50000, 1000000, 350000) | |
| # Conversion des entrées pour l'IA | |
| code_neuf = 1 if neuf_o_n == "OUI" else 0 | |
| code_etat = {"A rénover": 1, "Bon": 2, "Excellent": 3}[etat] | |
| code_dpe = {"A": 1, "B": 2, "C": 3, "D": 4, "E": 5, "F": 6, "G": 7}[dpe_label] | |
| code_park = 1 if parking else 0 | |
| code_balc = 1 if balcon else 0 | |
| code_jard = 1 if jardin else 0 | |
| code_asc = 1 if ascenseur else 0 | |
| # ========================================================== | |
| # 4. CALCULS ET RÉSULTATS | |
| # ========================================================== | |
| st.title("🚀 Prédilo IA : Expert Immobilier") | |
| st.markdown("---") | |
| # Prédiction en temps réel | |
| input_data = [[surface, distance, code_neuf, pieces, code_park, 0, code_balc, code_jard, code_asc, code_dpe, annee, code_etat]] | |
| estimation = model.predict(input_data)[0] | |
| # Affichage des métriques clés | |
| c1, c2, c3, c4 = st.columns(4) | |
| with c1: st.metric("Estimation IA", f"{estimation:,.0f} €") | |
| with c2: st.metric("Prix Vendeur", f"{prix_vendeur:,.0f} €") | |
| with c3: st.metric("Précision IA", f"{score_r2:.1%}") | |
| with c4: st.metric("Erreur (Test)", f"± {mae_test:,.0f} €") | |
| # Verdict visuel | |
| st.markdown('<div class="prediction-box">', unsafe_allow_html=True) | |
| if prix_vendeur < (estimation - mae_test): | |
| st.success("🎯 **VERDICT : EXCELLENTE AFFAIRE !**") | |
| st.write(f"Ce bien est sous-évalué d'environ **{estimation - prix_vendeur:,.0f} €**.") | |
| elif prix_vendeur > (estimation + mae_test): | |
| st.error("❌ **VERDICT : TROP CHER !**") | |
| st.write(f"Le prix est supérieur à la réalité du marché de **{prix_vendeur - estimation:,.0f} €**.") | |
| else: | |
| st.info("⚖️ **VERDICT : PRIX CORRECT !**") | |
| st.write("Le prix est parfaitement aligné avec les tendances actuelles.") | |
| st.markdown('</div>', unsafe_allow_html=True) | |
| # ========================================================== | |
| # 5. GRAPHIQUES PLOTLY | |
| # ========================================================== | |
| t1, t2 = st.tabs(["📊 Distribution du Marché", "📈 Relation Surface/Prix"]) | |
| with t1: | |
| fig_hist = px.histogram(df, x="prix", title="Où se situe votre bien ?", color_discrete_sequence=['#007bff']) | |
| fig_hist.add_vline(x=estimation, line_dash="dash", line_color="red", annotation_text="VOTRE ESTIMATION") | |
| st.plotly_chart(fig_hist, use_container_width=True) | |
| with t2: | |
| fig_scat = px.scatter(df, x="m2", y="prix", title="Prix en fonction de la Surface", labels={"m2": "Surface (m2)", "prix": "Prix (€)"}) | |
| fig_scat.add_trace(go.Scatter(x=[surface], y=[estimation], mode='markers', name='Votre Bien', marker=dict(size=15, color='red', symbol='star'))) | |
| st.plotly_chart(fig_scat, use_container_width=True) | |
| st.divider() | |
| st.caption("Développé avec ❤️ pour votre apprentissage de l'IA.") |