feat: house price prediction model integration

backend/app/ai_models/california.py

@@ -0,0 +1,33 @@
+from sklearn.datasets import fetch_california_housing
+from sklearn.linear_model import LinearRegression
+from sklearn.model_selection import train_test_split
+import joblib
+import pandas as pd
+
+# Load dataset
+data = fetch_california_housing()
+df = pd.DataFrame(data.data, columns=data.feature_names)
+df['target'] = data.target  # in 100k USD
+
+# Engineer features
+df['square_feet'] = df['AveRooms'] * 350  
+df['bedrooms'] = df['AveBedrms']
+df['bathrooms'] = df['AveRooms'] * 0.2 
+
+# Clean bathrooms
+df['bathrooms'] = df['bathrooms'].clip(lower=1)
+
+X = df[['square_feet', 'bedrooms', 'bathrooms']]
+y = df['target']
+
+# Train/test split
+X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
+
+# Train model
+model = LinearRegression()
+model.fit(X_train, y_train)
+
+# Need to be tested of course..: )
+
+# Save model
+joblib.dump(model, 'price_predictor.pkl')

backend/app/dtos/house/house_create_request.py

@@ -0,0 +1,34 @@
+from pydantic import BaseModel, Field
+
+
+class HouseCreateRequest(BaseModel):
+    address: str = Field(
+        ...,
+        min_length=1,
+        max_length=255,
+        description="House address",
+        examples=["123 Main St"],
+    )
+    city: str = Field(
+        ..., description="City where the house is located", examples=["Springfield"]
+    )
+    country: str = Field(
+        ..., description="Country where the house is located", examples=["USA"]
+    )
+    price: float = Field(..., description="Price of the house", examples=[250000.00])
+    description: str = Field(
+        ...,
+        description="Description of the house",
+        examples=["A beautiful 3-bedroom house"],
+    )
+    square_feet: float = Field(
+        ...,
+        description="Square footage of the house",
+        examples=[1500.00],
+    )
+    bedrooms: int = Field(
+        ..., description="Number of bedrooms in the house", examples=[3]
+    )
+    bathrooms: float = Field(
+        ..., description="Number of bathrooms in the house", examples=[2.5]
+    )

backend/app/dtos/house/house_create_response.py

@@ -0,0 +1,5 @@
+from pydantic import BaseModel
+
+
+class HouseCreateResponse(BaseModel):
+    id: str

backend/app/dtos/house/house_features.py

@@ -0,0 +1,8 @@
+from pydantic import BaseModel, Field
+from typing import Optional
+
+class HouseFeatures(BaseModel):
+    square_feet: float = Field(..., description="Total square feet of the house")
+    bedrooms: int = Field(..., description="Number of bedrooms")
+    bathrooms: float = Field(..., description="Number of bathrooms")
+    number_of_floors: Optional[int] = Field(default=None, description="Number of floors")

backend/app/dtos/house/house_predict_request.py

@@ -0,0 +1,7 @@
+from pydantic import BaseModel
+
+
+class HousePricePredictionRequest(BaseModel):
+    square_feet: float
+    bedrooms: int
+    bathrooms: float

backend/app/dtos/house/house_predict_response.py

@@ -0,0 +1,5 @@
+from pydantic import BaseModel
+
+
+class HousePricePredictionResponse(BaseModel):
+    predicted_price: float

backend/app/dtos/house/house_response.py

@@ -0,0 +1,10 @@
+from pydantic import BaseModel
+
+
+class HouseResponse(BaseModel):
+    id: str
+    description: str
+    address: str
+    city: str
+    country: str
+    price: float

backend/app/dtos/house/houses_list_response.py

@@ -0,0 +1,6 @@
+from pydantic import BaseModel
+
+from backend.app.dtos.house.house_response import HouseResponse
+
+class HousesListResponse(BaseModel):
+    houses: list[HouseResponse]

backend/app/dtos/owner/owner_detail_response.py

@@ -0,0 +1,7 @@
+from pydantic import BaseModel
+
+
+class OwnerDetailResponse(BaseModel):
+    id: str
+    user_id: str
+    email: str

backend/app/dtos/owner/owner_list_response.py

@@ -0,0 +1,10 @@
+from pydantic import BaseModel
+
+
+class OwnerResponse(BaseModel):
+    id: str
+    user_id: str
+
+
+class OwnerListResponse(BaseModel):
+    owners: list[OwnerResponse]

backend/app/dtos/user/user_list_response.py

@@ -0,0 +1,6 @@
+from pydantic import BaseModel
+
+from backend.app.dtos.user.user_response import UserResponse
+
+class UserListResponse(BaseModel):
+    users: list[UserResponse]

backend/app/dtos/user/user_response.py

@@ -0,0 +1,6 @@
+from pydantic import BaseModel
+
+
+class UserResponse(BaseModel):
+    id: str
+    email: str

backend/app/models/house.py

@@ -1,6 +1,11 @@
+from typing import Optional, TYPE_CHECKING  
 from uuid import UUID, uuid4
 
-from sqlmodel import Field, SQLModel
+from sqlmodel import Field, Relationship, SQLModel
+
+if TYPE_CHECKING:
+    from backend.app.models.owner import Owner
+
 
 
 class House(SQLModel, table=True):
@@ -14,3 +19,4 @@ class House(SQLModel, table=True):
     square_feet: float = Field()
     bedrooms: int = Field()
     bathrooms: float = Field()
+    owner: Optional["Owner"] = Relationship(back_populates="houses")

backend/app/models/owner.py

@@ -1,8 +1,17 @@
+from typing import Optional, TYPE_CHECKING
 from uuid import UUID, uuid4
 
-from sqlmodel import Field, SQLModel
+from sqlmodel import Field, Relationship, SQLModel
+
+if TYPE_CHECKING:
+    from backend.app.models.house import House
+    from backend.app.models.user import User
 
 
 class Owner(SQLModel, table=True):
     id: UUID = Field(default_factory=uuid4, primary_key=True)
     user_id: UUID = Field(foreign_key="user.id", unique=True)
+    
+    # Relationship
+    houses: list["House"] = Relationship(back_populates="owner")
+    user: Optional["User"] = Relationship(back_populates="owner")

backend/app/models/user.py

@@ -1,9 +1,16 @@
+from typing import Optional, TYPE_CHECKING
 from uuid import UUID, uuid4
 
-from sqlmodel import Field, SQLModel
+from sqlmodel import Field, Relationship, SQLModel
+
+if TYPE_CHECKING:
+    from backend.app.models.owner import Owner
 
 
 class User(SQLModel, table=True):
     id: UUID = Field(default_factory=lambda: uuid4(), primary_key=True)
     email: str = Field(unique=True, nullable=False)
     password_hash: str = Field(nullable=False)
+
+    # Relationships
+    owner: Optional["Owner"] = Relationship(back_populates="user")

backend/app/repositories/house_repository.py

@@ -8,7 +8,6 @@ from sqlmodel import asc, desc, select
 from ..models.house import House
 from ..providers.db_provider import get_session
 
-
 class HouseRepository:
     def __init__(self, session: Annotated[AsyncSession, Depends(get_session)]) -> None:
         self.session = session
@@ -34,6 +33,11 @@ class HouseRepository:
         result = await self.session.execute(statement)
         return result.scalar_one_or_none()
 
+    async def get_by_user_id(self, user_id: UUID):
+        statement = select(House).where(House.owner_user_id == user_id)
+        result = await self.session.execute(statement)
+        return result.scalars().all()
+    
     async def save(self, house: House) -> None:
         """
         Save a house to the database. If a house with that ID already exists, do an upsert.

backend/app/repositories/owner_repository.py

@@ -5,6 +5,9 @@ from fastapi import Depends
 from sqlalchemy.ext.asyncio.session import AsyncSession
 from sqlmodel import select
 
+from backend.app.models.house import House
+from backend.app.models.user import User
+
 from ..models.owner import Owner
 from ..providers.db_provider import get_session
 
@@ -27,7 +30,26 @@ class OwnerRepository:
         statement = select(Owner).where(Owner.user_id == user_id)
         result = await self.session.execute(statement)
         return result.scalar_one_or_none()
+    
+    async def get_details_by_house_id(self, house_id: UUID):
+        statement = (
+            select(Owner, User)
+            .join(User, Owner.user_id == User.id)
+            .join(House, House.owner_user_id == Owner.user_id)
+            .where(House.id == house_id)
+        )
+        result = await self.session.execute(statement)
+        row = result.first()  
+       
+        if row:
+            owner, user = row
+            return {
+                "owner": owner,
+                "user": user
+            }
 
+        return None
+    
     async def save(self, owner: Owner) -> None:
         """
         Save a owner to the database. If an owner with that ID already exists, do an upsert.

backend/app/routers/houses.py

@@ -2,15 +2,19 @@ from typing import Annotated, Literal
 
 from fastapi import APIRouter, Depends
 
-from ..dtos.house_create_request import HouseCreateRequest
+from backend.app.dtos.house.house_create_request import HouseCreateRequest
-from ..dtos.house_create_response import HouseCreateResponse
+from backend.app.dtos.house.house_create_response import HouseCreateResponse
-from ..dtos.houses_list_response import HouseResponse, HousesListResponse
+from backend.app.dtos.house.house_features import HouseFeatures
+from backend.app.dtos.house.house_predict_request import HousePricePredictionRequest
+from backend.app.dtos.house.house_predict_response import HousePricePredictionResponse
+from backend.app.dtos.house.house_response import HouseResponse
+from backend.app.dtos.house.houses_list_response import HousesListResponse
 from ..models.house import House
 from ..models.owner import Owner
 from ..providers.auth_provider import AuthContext
 from ..repositories.house_repository import HouseRepository
 from ..repositories.owner_repository import OwnerRepository
-
+from ..services.house_price_predictor import HousePricePredictor
 router = APIRouter()
 
 
@@ -69,4 +73,22 @@ async def get_all_houses(
         for house in all_houses
     ]
 
+
     return HousesListResponse(houses=house_responses)
+
+@router.post("/predict-price")
+async def predict_house_price(
+        body: HouseFeatures,
+        price_predictor: Annotated[HousePricePredictor, Depends()],
+    ) -> HousePricePredictionResponse:
+        """
+        Predict the price of a house based on its features.
+        """
+        predicted_price = await price_predictor.predict_california(
+            square_feet=body.square_feet,
+            bedrooms=body.bedrooms,
+            bathrooms=body.bathrooms
+        )
+        
+        return HousePricePredictionResponse(predicted_price=predicted_price)  
+

backend/app/routers/owners.py

@@ -1,15 +1,14 @@
 from typing import Annotated
 
-from fastapi import APIRouter, Depends
+from fastapi import APIRouter, Depends, HTTPException
 
-from ..dtos.owner_detail_response import OwnerDetailResponse
+from ..dtos.owner.owner_detail_response import OwnerDetailResponse
-from ..dtos.owner_list_response import OwnerListResponse, OwnerResponse
+from ..dtos.owner.owner_list_response import OwnerListResponse, OwnerResponse
 from ..repositories.owner_repository import OwnerRepository
 from ..repositories.user_repository import UserRepository
 
 router = APIRouter()
 
-
 @router.get("")
 async def get_owners(
     owner_repository: Annotated[OwnerRepository, Depends()],
@@ -22,7 +21,6 @@ async def get_owners(
 
     return OwnerListResponse(owners=owners_response)
 
-
 @router.get("/{id}")
 async def get_owner(
     id: str,
@@ -35,3 +33,22 @@ async def get_owner(
     return OwnerDetailResponse(
         id=str(owner.id), user_id=str(owner.user_id), email=user.email
     )
+
+@router.get("/byhouse/{house_id}")
+async def get_owner_by_house_id(
+    house_id: str,
+    owner_repository: Annotated[OwnerRepository, Depends()],
+) -> OwnerDetailResponse:
+    result = await owner_repository.get_details_by_house_id(house_id)
+    
+    if result is None:
+        raise HTTPException(status_code=404, detail="House or owner not found")
+        
+    owner = result["owner"]
+    user = result["user"]
+    
+    return OwnerDetailResponse(
+        id=str(owner.id),
+        user_id=str(owner.user_id),
+        email=str(user.email)
+    )

backend/app/services/house_price_predictor.py

@@ -1,16 +1,17 @@
+import os
+import joblib
+import numpy as np
+
+from backend.app.dtos.house.house_features import HouseFeatures
+
 class HousePricePredictor:
     """
     Mock ML model that predicts house prices.
     In a real scenario, this would load a trained model.
     """
-
+    def __init__(self):
-    async def predict(
+        self.model = joblib.load("backend/app/ai_models/price_predictor.pkl")
-        self, square_feet: float, bedrooms: int, bathrooms: float
+        
-    ) -> float:
+    def predict(self, features: HouseFeatures) -> float:
-        base_price = square_feet * 200
+        X = np.array([[features.square_feet, features.bedrooms, features.bathrooms]])
-        bedroom_value = bedrooms * 25000
+        return self.model.predict(X)[0] * 100000 
-        bathroom_value = bathrooms * 15000
-
-        predicted_price = base_price + bedroom_value + bathroom_value
-
-        return predicted_price

backend/requirements.txt

@@ -8,3 +8,5 @@ python-dotenv
 pg8000
 asyncpg
 greenlet
+botbuilder-core 
+openai