Upload 30 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+data/online_retail_cleaned.csv filter=lfs diff=lfs merge=lfs -text
+data/OnlineRetail.csv filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,11 @@
+FROM python:3.11-slim
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+EXPOSE 7860
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]

app/database.py

@@ -0,0 +1,15 @@
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+import os
+
+DATABASE_URL = os.getenv("DATABASE_URL", "postgresql://postgres:samadhi@localhost:5432/ecommerce_db")
+
+engine = create_engine(DATABASE_URL)
+SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
+
+def get_db():
+    db = SessionLocal()
+    try:
+        yield db
+    finally:
+        db.close()

app/main.py

@@ -0,0 +1,101 @@
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from app.database import engine
+from app.ml_models import load_models
+from app.routes import customers, predictions
+from sqlalchemy import text
+
+# load ML models on startup
+models_loaded = load_models()
+
+# create fastapi app
+app = FastAPI(
+    title="E-Commerce Customer Intelligence API",
+    description="""
+    This API provides customer insights from your e-commerce data.
+    
+    ## Features
+    * **Customer Segmentation** - Predict customer segments using K-Means
+    * **CLV Prediction** - Predict Customer Lifetime Value
+    * **Customer Data** - Access customer information from database
+    
+    ## Models
+    * K-Means Clustering (4 segments)
+    * Random Forest CLV Predictor (95% accuracy)
+    """,
+    version="1.0.0",
+    contact={
+        "name": "Your Name",
+        "email": "your.email@example.com",
+    },
+)
+
+#add CORS middleware (allows frontend apps to call your API)
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # In production, specify actual domains
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+#include routers
+app.include_router(customers.router)
+app.include_router(predictions.router)
+
+@app.get("/", tags=["Root"])
+def root():
+    """Welcome endpoint"""
+    return {
+        "message": "Welcome to E-Commerce Customer Intelligence API",
+        "docs": "/docs",
+        "version": "1.0.0",
+        "status": "operational"
+    }
+
+@app.get("/health", tags=["Health"])
+def health_check():
+    """Check if API and database are working"""
+    db_status = "unknown"
+    
+    try:
+        #test database connection
+        with engine.connect() as conn:
+            conn.execute(text("SELECT 1"))
+            db_status = "connected"
+    except Exception as e:
+        db_status = f"error: {str(e)}"
+    
+    return {
+        "status": "healthy",
+        "models_loaded": {
+            "kmeans": models_loaded,
+            "clv": models_loaded
+        },
+        "database": db_status,
+        "timestamp": "2024-01-01T00:00:00Z"
+    }
+
+@app.get("/info", tags=["Info"])
+def api_info():
+    """Get API information and available endpoints"""
+    return {
+        "name": "E-Commerce Customer Intelligence API",
+        "version": "1.0.0",
+        "endpoints": {
+            "GET /": "Welcome message",
+            "GET /health": "Health check",
+            "GET /info": "This information",
+            "GET /customers": "List all customers",
+            "GET /customers/{id}": "Get customer by ID",
+            "GET /customers/{id}/transactions": "Get customer transactions",
+            "POST /predict/segment": "Predict customer segment from RFM",
+            "POST /predict/clv": "Predict Customer Lifetime Value"
+        },
+        "models": {
+            "customer_segmentation": "K-Means (4 clusters)",
+            "clv_prediction": "Random Forest (95% accuracy)"
+        }
+    }
+
+#run with: uvicorn app.main:app --reload

app/ml_models.py

@@ -0,0 +1,137 @@
+import joblib
+import os
+import pandas as pd
+import numpy as np
+
+#global variables for models
+kmeans_model = None
+clv_model = None
+scaler = None
+
+def load_models():
+    """Load all trained ML models"""
+    global kmeans_model, clv_model, scaler
+
+    
+    current_dir = os.path.dirname(os.path.abspath(__file__))
+    project_root = os.path.dirname(current_dir) 
+    models_path = os.path.join(project_root, "models")
+    
+    print(f"Looking for models in: {models_path}")
+    
+    if not os.path.exists(models_path):
+        print(f"Models folder not found at: {models_path}")
+        return False
+
+    try:
+        #load K-Means model for customer segmentation
+        kmeans_path = os.path.join(models_path, "kmeans_model.pkl")
+        if os.path.exists(kmeans_path):
+            kmeans_model = joblib.load(kmeans_path)
+            print("K-Means model loaded")
+        else:
+            print(f"File not found: {kmeans_path}")
+            kmeans_model = None
+    except Exception as e:
+        print(f"Could not load K-Means model: {e}")
+        kmeans_model = None
+
+    try:
+        #load CLV prediction model (FIX 2: Changed 'csv' to 'clv')
+        clv_path = os.path.join(models_path, "clv_model.pkl")
+        if os.path.exists(clv_path):
+            clv_model = joblib.load(clv_path)
+            print("CLV model loaded")
+        else:
+            print(f"File not found: {clv_path}")
+            clv_model = None
+    except Exception as e:
+        print(f"Could not load CLV model: {e}")
+        clv_model = None
+
+    try:
+        #load scaler
+        scaler_path = os.path.join(models_path, "scaler.pkl")
+        if os.path.exists(scaler_path):
+            scaler = joblib.load(scaler_path)
+            print("Scaler loaded")
+        else:
+            print(f"File not found: {scaler_path}")
+            scaler = None
+    except Exception as e:
+        print(f"Could not load scaler: {e}")
+        scaler = None
+
+    return kmeans_model is not None or clv_model is not None
+
+def predict_segment(recency, frequency, monetary):
+    """Predict customer segment using K-Means model"""
+    if kmeans_model is None or scaler is None:
+        return {"error": "Models not loaded"}
+    
+    #create dataframe with correct feature order
+    customer_data = pd.DataFrame({
+        'Recency': [recency],
+        'Frequency': [frequency],
+        'Monetary': [monetary]
+    })
+
+    #scale the features
+    scaled_data = scaler.transform(customer_data)
+
+    #predict cluster
+    cluster = kmeans_model.predict(scaled_data)[0]
+
+    #map cluster to segment name
+    segment_map = {
+        0: "At-Risk Customers",   # 2,396 customers - high recency, low frequency
+        1: "VIP Customers",       # 1,024 customers - very high recency (lost customers)
+        2: "Loyal Regulars",      # 145 customers - low recency, high frequency (YOUR BEST!)
+        3: "New/Occasional"       # 723 customers - medium recency, medium frequency
+    }
+
+    return {
+        "cluster": int(cluster),
+        "segment": segment_map.get(cluster, "Unknown")
+    }
+
+def predict_clv(features_dict):
+    """
+    Predict Customer Lifetime Value
+    features_dict should contain all 9 features
+    """
+    if clv_model is None:
+        return {"error": "CLV model not loaded"}
+    
+    #expected feature order from your notebook
+    feature_columns = [
+        'frequency', 'recency', 'avg_quantity', 'avg_unit_price',
+        'avg_transaction', 'lifespan_days', 'avg_days_between_purchases',
+        'purchases_per_month', 'total_quantity'
+    ]
+    
+    #create list of features in correct order
+    features = []
+    for col in feature_columns:
+        features.append(features_dict.get(col, 0))
+
+    #reshape for prediction (1 sample with 9 features)
+    features_array = np.array(features).reshape(1, -1)
+
+    #predict
+    prediction = clv_model.predict(features_array)[0]
+
+    #determine value category (adjust bins based on model)
+    if prediction < 500:
+        category = "Low Value"
+    elif prediction < 2000:
+        category = "Medium Value"
+    elif prediction < 5000:
+        category = "High Value"
+    else:
+        category = "VIP"
+
+    return {
+        "predicted_clv": round(prediction, 2),
+        "value_category": category
+    }

app/models.py

@@ -0,0 +1,61 @@
+from pydantic import BaseModel, Field
+from typing import Optional, List
+
+#request models
+class CustomerFeatures(BaseModel):
+    """Features for customer segmentation"""
+    recency: int = Field(..., description="Days since last purchase", ge=0)
+    frequency: int = Field(..., description="Number of purchases", ge=1)
+    monetary: float = Field(..., description="Total amount spent", ge=0)
+
+class CLVFeatures(BaseModel):
+    """Features for CLV prediction"""
+    frequency: int = Field(..., ge=1)
+    recency: int = Field(..., ge=0)
+    avg_quantity: float = Field(..., ge=0)
+    avg_unit_price: float = Field(..., ge=0)
+    avg_transaction: float = Field(..., ge=0)
+    lifespan_days: int = Field(..., ge=0)
+    avg_days_between_purchases: float = Field(..., ge=0)
+    purchases_per_month: float = Field(..., ge=0)
+    total_quantity: int = Field(..., ge=0)
+
+    class Config:
+        schema_extra = {
+            "example": {
+                "frequency": 12,
+                "recency": 7,
+                "avg_quantity": 3.5,
+                "avg_unit_price": 25.0,
+                "avg_transaction": 87.5,
+                "lifespan_days": 180,
+                "avg_days_between_purchases": 15,
+                "purchases_per_month": 2,
+                "total_quantity": 42
+            }
+        }
+
+#response models
+class SegmentResponse(BaseModel):
+    customer_id: Optional[int] = None
+    cluster: int
+    segment: str
+    recency: Optional[int] = None
+    frequency: Optional[int] = None
+    monetary: Optional[float] = None
+
+class CLVResponse(BaseModel):
+    predicted_clv: float
+    value_category: str
+
+class CustomerInfo(BaseModel):
+    customer_id: int
+    segment: str
+    value_category: str
+    total_orders: Optional[int] = None
+    total_revenue: Optional[float] = None
+
+class HealthResponse(BaseModel):
+    status: str
+    models_loaded: dict
+    database: str

app/routes/customers.py

@@ -0,0 +1,119 @@
+from fastapi import APIRouter, Depends, HTTPException
+from sqlalchemy.orm import Session
+from sqlalchemy import text
+from typing import List
+from app.database import get_db
+from app.models import CustomerInfo
+
+router = APIRouter(prefix="/customers", tags=["Customers"])
+
+@router.get("/", response_model=List[CustomerInfo])
+def get_all_customers(
+    skip: int = 0, 
+    limit: int = 100, 
+    db: Session = Depends(get_db)
+):
+    """Get list of all customers with their segments"""
+    
+    query = text("""
+        SELECT 
+            t."CustomerID" as customer_id,
+            s."Segment" as segment,
+            v."value_category" as value_category,
+            COUNT(DISTINCT t."InvoiceNo") as total_orders,
+            SUM(t."TotalPrice") as total_revenue
+        FROM transactions t
+        LEFT JOIN customer_segments s ON t."CustomerID" = s."CustomerID"
+        LEFT JOIN clv_predictions v ON t."CustomerID" = v."CustomerID"
+        GROUP BY t."CustomerID", s."Segment", v."value_category"
+        ORDER BY total_revenue DESC
+        LIMIT :limit OFFSET :skip
+    """)
+    
+    result = db.execute(query, {"limit": limit, "skip": skip}).fetchall()
+    
+    customers = []
+    for row in result:
+        customers.append({
+            "customer_id": row[0],
+            "segment": row[1] or "Unknown",
+            "value_category": row[2] or "Unknown",
+            "total_orders": row[3],
+            "total_revenue": float(row[4]) if row[4] else 0
+        })
+    
+    return customers
+
+@router.get("/{customer_id}", response_model=CustomerInfo)
+def get_customer_by_id(customer_id: int, db: Session = Depends(get_db)):
+    """Get customer details by ID"""
+    
+    query = text("""
+        SELECT 
+            t."CustomerID" as customer_id,
+            s."Segment" as segment,
+            v."value_category" as value_category,
+            COUNT(DISTINCT t."InvoiceNo") as total_orders,
+            SUM(t."TotalPrice") as total_revenue
+        FROM transactions t
+        LEFT JOIN customer_segments s ON t."CustomerID" = s."CustomerID"
+        LEFT JOIN clv_predictions v ON t."CustomerID" = v."CustomerID"
+        WHERE t."CustomerID" = :customer_id
+        GROUP BY t."CustomerID", s."Segment", v."value_category"
+    """)
+    
+    result = db.execute(query, {"customer_id": customer_id}).fetchone()
+    
+    if not result:
+        raise HTTPException(status_code=404, detail="Customer not found")
+    
+    return {
+        "customer_id": result[0],
+        "segment": result[1] or "Unknown",
+        "value_category": result[2] or "Unknown",
+        "total_orders": result[3],
+        "total_revenue": float(result[4]) if result[4] else 0
+    }
+
+@router.get("/{customer_id}/transactions")
+def get_customer_transactions(
+    customer_id: int, 
+    limit: int = 50, 
+    db: Session = Depends(get_db)
+):
+    """Get transaction history for a customer"""
+    
+    query = text("""
+        SELECT 
+            "InvoiceNo",
+            "InvoiceDate",
+            "StockCode",
+            "Description",
+            "Quantity",
+            "UnitPrice",
+            "TotalPrice"
+        FROM transactions
+        WHERE "CustomerID" = :customer_id
+        ORDER BY "InvoiceDate" DESC
+        LIMIT :limit
+    """)
+    
+    result = db.execute(query, {"customer_id": customer_id, "limit": limit}).fetchall()
+    
+    transactions = []
+    for row in result:
+        transactions.append({
+            "invoice_no": row[0],
+            "date": str(row[1]),
+            "stock_code": row[2],
+            "description": row[3],
+            "quantity": row[4],
+            "unit_price": float(row[5]),
+            "total_price": float(row[6])
+        })
+    
+    return {
+        "customer_id": customer_id,
+        "transaction_count": len(transactions),
+        "transactions": transactions
+    }

app/routes/predictions.py

@@ -0,0 +1,50 @@
+from fastapi import APIRouter, HTTPException
+from app.models import CustomerFeatures, CLVFeatures, SegmentResponse, CLVResponse
+from app.ml_models import predict_segment, predict_clv
+
+router = APIRouter(prefix="/predict", tags=["Predictions"])
+
+@router.post("/segment", response_model=SegmentResponse)
+def segment_customer(features: CustomerFeatures):
+    """Predict customer segment based on RFM features"""
+    
+    result = predict_segment(
+        features.recency,
+        features.frequency,
+        features.monetary
+    )
+    
+    if "error" in result:
+        raise HTTPException(status_code=500, detail=result["error"])
+    
+    return {
+        "cluster": result["cluster"],
+        "segment": result["segment"],
+        "recency": features.recency,
+        "frequency": features.frequency,
+        "monetary": features.monetary
+    }
+
+@router.post("/clv", response_model=CLVResponse)
+def predict_customer_value(features: CLVFeatures):
+    """Predict Customer Lifetime Value"""
+    
+    # convert features to dictionary
+    features_dict = features.dict()
+    
+    result = predict_clv(features_dict)
+    
+    if "error" in result:
+        raise HTTPException(status_code=500, detail=result["error"])
+    
+    return result
+
+@router.get("/segment/{customer_id}")
+def get_customer_segment_from_db(customer_id: int):
+    """
+    Get pre-computed segment for a customer from database
+    This endpoint connects to PostgreSQL to get the segment
+    """
+    # this will be implemented with database dependency
+    # for now, return a placeholder
+    return {"customer_id": customer_id, "message": "To be implemented with database"}

app/upload_to_neon.py

@@ -0,0 +1,31 @@
+import pandas as pd
+from sqlalchemy import create_engine
+
+DATABASE_URL = "postgresql://neondb_owner:npg_meVF3arI6qWv@ep-gentle-frog-a170mdjt-pooler.ap-southeast-1.aws.neon.tech/neondb?sslmode=require&channel_binding=require"
+
+print("Connecting to Neon...")
+engine = create_engine(DATABASE_URL)
+
+print("Loading CSV files...")
+
+transactions = pd.read_csv('C:/Users/User/Desktop/ecommerce/hf_deploy/data/online_retail_cleaned.csv')
+segments = pd.read_csv('C:/Users/User/Desktop/ecommerce/hf_deploy/data/customer_segments.csv')
+clv = pd.read_csv('C:/Users/User/Desktop/ecommerce/hf_deploy/data/clv_predictions.csv')
+rfm = pd.read_csv('C:/Users/User/Desktop/ecommerce/hf_deploy/data/features_rfm.csv')
+daily = pd.read_csv('C:/Users/User/Desktop/ecommerce/hf_deploy/data/features_daily_sales.csv')
+
+print(f"transactions: {len(transactions)} rows")
+print(f"segments: {len(segments)} rows")
+print(f"clv: {len(clv)} rows")
+print(f"rfm: {len(rfm)} rows")
+print(f"daily: {len(daily)} rows")
+
+print("Uploading to Neon...")
+
+transactions.to_sql('transactions', engine, if_exists='replace', index=False, chunksize=5000)
+segments.to_sql('customer_segments', engine, if_exists='replace', index=False)
+clv.to_sql('clv_predictions', engine, if_exists='replace', index=False)
+rfm.to_sql('customer_rfm', engine, if_exists='replace', index=False)
+daily.to_sql('daily_sales', engine, if_exists='replace', index=False)
+
+print("ALL DATA UPLOADED!")

data/OnlineRetail.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d07aec9960083af2339975a3f9d3b26313b342dcd9f86cce0b919b1cde639a44
+size 45580638

data/features_daily_sales.csv

@@ -0,0 +1,306 @@
+Date,Revenue,Order,Quantity,DayOfWeek,Month,Year
+2010-12-01,43616.49,120,23351,Wednesday,12,2010
+2010-12-02,43252.53,136,27774,Thursday,12,2010
+2010-12-03,22607.71,56,11719,Friday,12,2010
+2010-12-05,31771.600000000002,87,16449,Sunday,12,2010
+2010-12-06,28262.440000000002,94,15655,Monday,12,2010
+2010-12-07,28955.37,71,12938,Tuesday,12,2010
+2010-12-08,39248.82,111,21573,Wednesday,12,2010
+2010-12-09,32523.260000000002,96,15831,Thursday,12,2010
+2010-12-10,30671.88,72,15779,Friday,12,2010
+2010-12-12,17305.77,43,10599,Sunday,12,2010
+2010-12-13,27642.68,64,15335,Monday,12,2010
+2010-12-14,28134.3,80,17496,Tuesday,12,2010
+2010-12-15,28533.32,70,17868,Wednesday,12,2010
+2010-12-16,43871.88,113,26413,Thursday,12,2010
+2010-12-17,20046.56,55,11407,Friday,12,2010
+2010-12-19,7417.39,18,3739,Sunday,12,2010
+2010-12-20,19621.96,50,13445,Monday,12,2010
+2010-12-21,15951.66,27,11068,Tuesday,12,2010
+2010-12-22,4886.52,14,3074,Wednesday,12,2010
+2010-12-23,5648.97,17,3222,Thursday,12,2010
+2011-01-04,12125.460000000001,35,6959,Tuesday,1,2011
+2011-01-05,25783.23,47,17291,Wednesday,1,2011
+2011-01-06,33340.19,46,21298,Thursday,1,2011
+2011-01-07,23797.79,47,15122,Friday,1,2011
+2011-01-09,15778.2,48,8202,Sunday,1,2011
+2011-01-10,15346.83,34,9387,Monday,1,2011
+2011-01-11,20382.24,50,11653,Tuesday,1,2011
+2011-01-12,17153.94,42,7899,Wednesday,1,2011
+2011-01-13,15171.31,41,8312,Thursday,1,2011
+2011-01-14,35555.22,38,20708,Friday,1,2011
+2011-01-16,7242.06,25,4204,Sunday,1,2011
+2011-01-17,16597.35,46,8714,Monday,1,2011
+2011-01-18,10405.51,31,6403,Tuesday,1,2011
+2011-01-19,21646.78,33,15902,Wednesday,1,2011
+2011-01-20,15244.73,34,8185,Thursday,1,2011
+2011-01-21,24279.76,34,12697,Friday,1,2011
+2011-01-23,10400.25,27,5241,Sunday,1,2011
+2011-01-24,17632.37,44,9601,Monday,1,2011
+2011-01-25,24119.3,59,14519,Tuesday,1,2011
+2011-01-26,17490.82,53,10455,Wednesday,1,2011
+2011-01-27,22157.03,50,10950,Thursday,1,2011
+2011-01-28,17273.5,37,9174,Friday,1,2011
+2011-01-30,6615.75,24,3431,Sunday,1,2011
+2011-01-31,18818.08,57,12426,Monday,1,2011
+2011-02-01,26376.5,58,15013,Tuesday,2,2011
+2011-02-02,17145.23,56,8931,Wednesday,2,2011
+2011-02-03,20556.8,45,14954,Thursday,2,2011
+2011-02-04,18114.31,45,11264,Friday,2,2011
+2011-02-06,3457.11,11,2048,Sunday,2,2011
+2011-02-07,13682.41,37,6362,Monday,2,2011
+2011-02-08,14168.03,38,8693,Tuesday,2,2011
+2011-02-09,11920.78,23,6145,Wednesday,2,2011
+2011-02-10,14198.51,40,11784,Thursday,2,2011
+2011-02-11,16343.460000000001,37,7179,Friday,2,2011
+2011-02-13,5713.63,20,2756,Sunday,2,2011
+2011-02-14,21884.63,35,11287,Monday,2,2011
+2011-02-15,37784.65,53,22107,Tuesday,2,2011
+2011-02-16,23499.22,62,15768,Wednesday,2,2011
+2011-02-17,17674.79,55,12762,Thursday,2,2011
+2011-02-18,14463.06,35,8642,Friday,2,2011
+2011-02-20,9624.69,26,5353,Sunday,2,2011
+2011-02-21,32801.73,33,20432,Monday,2,2011
+2011-02-22,25957.56,49,17024,Tuesday,2,2011
+2011-02-23,19348.07,53,12164,Wednesday,2,2011
+2011-02-24,21472.89,53,11563,Thursday,2,2011
+2011-02-25,17082.29,44,9835,Friday,2,2011
+2011-02-27,9526.5,33,4870,Sunday,2,2011
+2011-02-28,15107.02,51,8420,Monday,2,2011
+2011-03-01,22407.87,56,11478,Tuesday,3,2011
+2011-03-02,17044.49,41,8463,Wednesday,3,2011
+2011-03-03,30385.05,46,18851,Thursday,3,2011
+2011-03-04,17322.47,45,12942,Friday,3,2011
+2011-03-06,9997.42,26,5048,Sunday,3,2011
+2011-03-07,20285.68,58,11112,Monday,3,2011
+2011-03-08,22230.61,44,13841,Tuesday,3,2011
+2011-03-09,19187.87,51,10914,Wednesday,3,2011
+2011-03-10,24431.72,52,15223,Thursday,3,2011
+2011-03-11,17521.66,46,9248,Friday,3,2011
+2011-03-13,4148.12,16,2749,Sunday,3,2011
+2011-03-14,26009.25,48,16757,Monday,3,2011
+2011-03-15,14936.43,41,7937,Tuesday,3,2011
+2011-03-16,21820.66,50,12907,Wednesday,3,2011
+2011-03-17,25098.94,56,14852,Thursday,3,2011
+2011-03-18,23398.74,52,13258,Friday,3,2011
+2011-03-20,20084.8,57,13630,Sunday,3,2011
+2011-03-21,16051.05,48,9456,Monday,3,2011
+2011-03-22,19064.29,41,13503,Tuesday,3,2011
+2011-03-23,21578.8,59,13934,Wednesday,3,2011
+2011-03-24,28728.34,68,16136,Thursday,3,2011
+2011-03-25,21789.43,51,12002,Friday,3,2011
+2011-03-27,9224.4,31,4529,Sunday,3,2011
+2011-03-28,18887.66,58,11465,Monday,3,2011
+2011-03-29,35078.52,50,22903,Tuesday,3,2011
+2011-03-30,29587.420000000002,66,20143,Wednesday,3,2011
+2011-03-31,25688.61,59,14984,Thursday,3,2011
+2011-04-01,23670.98,67,17449,Friday,4,2011
+2011-04-03,6918.5,19,5667,Sunday,4,2011
+2011-04-04,23328.83,53,12867,Monday,4,2011
+2011-04-05,22199.07,44,14011,Tuesday,4,2011
+2011-04-06,12722.32,37,8025,Wednesday,4,2011
+2011-04-07,16827.77,61,10100,Thursday,4,2011
+2011-04-08,20852.85,64,12297,Friday,4,2011
+2011-04-10,9913.98,32,5632,Sunday,4,2011
+2011-04-11,20184.14,59,13133,Monday,4,2011
+2011-04-12,24023.96,60,14607,Tuesday,4,2011
+2011-04-13,23365.06,59,17713,Wednesday,4,2011
+2011-04-14,34726.38,82,18238,Thursday,4,2011
+2011-04-15,18021.481,41,11100,Friday,4,2011
+2011-04-17,12725.5,42,8183,Sunday,4,2011
+2011-04-18,22177.66,61,17224,Monday,4,2011
+2011-04-19,17882.84,52,12461,Tuesday,4,2011
+2011-04-20,25718.95,62,17856,Wednesday,4,2011
+2011-04-21,27804.47,71,17081,Thursday,4,2011
+2011-04-26,20666.41,58,14070,Tuesday,4,2011
+2011-04-27,21950.7,60,17539,Wednesday,4,2011
+2011-04-28,21323.29,55,13122,Thursday,4,2011
+2011-05-01,6973.66,18,3819,Sunday,5,2011
+2011-05-03,20767.66,58,11145,Tuesday,5,2011
+2011-05-04,27532.0,62,17381,Wednesday,5,2011
+2011-05-05,25232.670000000002,82,16231,Thursday,5,2011
+2011-05-06,30786.7,77,18287,Friday,5,2011
+2011-05-08,18867.4,63,10702,Sunday,5,2011
+2011-05-09,21624.41,63,11682,Monday,5,2011
+2011-05-10,29870.7,70,16399,Tuesday,5,2011
+2011-05-11,30956.2,72,17091,Wednesday,5,2011
+2011-05-12,57410.56,83,36597,Thursday,5,2011
+2011-05-13,25885.83,67,13294,Friday,5,2011
+2011-05-15,9680.05,30,4771,Sunday,5,2011
+2011-05-16,32518.7,64,14655,Monday,5,2011
+2011-05-17,45230.090000000004,72,25005,Tuesday,5,2011
+2011-05-18,32882.87,71,18637,Wednesday,5,2011
+2011-05-19,31553.510000000002,92,17180,Thursday,5,2011
+2011-05-20,25489.66,69,15129,Friday,5,2011
+2011-05-22,22531.09,61,12549,Sunday,5,2011
+2011-05-23,27689.53,62,14831,Monday,5,2011
+2011-05-24,20168.600000000002,60,10666,Tuesday,5,2011
+2011-05-25,20492.93,59,11481,Wednesday,5,2011
+2011-05-26,28310.82,61,14470,Thursday,5,2011
+2011-05-27,18067.32,55,10539,Friday,5,2011
+2011-05-29,7394.3,24,4124,Sunday,5,2011
+2011-05-31,18194.65,51,10607,Tuesday,5,2011
+2011-06-01,15390.89,37,9580,Wednesday,6,2011
+2011-06-02,28104.56,42,13714,Thursday,6,2011
+2011-06-03,12589.22,39,6792,Friday,6,2011
+2011-06-05,25639.54,67,13501,Sunday,6,2011
+2011-06-06,16290.98,56,9142,Monday,6,2011
+2011-06-07,22908.74,70,15422,Tuesday,6,2011
+2011-06-08,30192.13,82,18833,Wednesday,6,2011
+2011-06-09,26422.38,81,23594,Thursday,6,2011
+2011-06-10,19275.12,43,9773,Friday,6,2011
+2011-06-12,12472.210000000001,38,9488,Sunday,6,2011
+2011-06-13,18498.71,52,10304,Monday,6,2011
+2011-06-14,22755.63,51,10896,Tuesday,6,2011
+2011-06-15,43085.54,50,29050,Wednesday,6,2011
+2011-06-16,31396.97,81,19373,Thursday,6,2011
+2011-06-17,19059.23,46,11349,Friday,6,2011
+2011-06-19,22442.18,60,15081,Sunday,6,2011
+2011-06-20,26136.73,59,14873,Monday,6,2011
+2011-06-21,19850.66,48,13127,Tuesday,6,2011
+2011-06-22,21170.420000000002,57,15291,Wednesday,6,2011
+2011-06-23,22556.81,69,14030,Thursday,6,2011
+2011-06-24,16021.710000000001,45,9067,Friday,6,2011
+2011-06-26,7082.49,26,3786,Sunday,6,2011
+2011-06-27,13412.2,35,9163,Monday,6,2011
+2011-06-28,30532.37,49,20926,Tuesday,6,2011
+2011-06-29,12570.630000000001,39,7469,Wednesday,6,2011
+2011-06-30,25433.96,64,14913,Thursday,6,2011
+2011-07-01,12189.29,42,7454,Friday,7,2011
+2011-07-03,6032.39,25,3117,Sunday,7,2011
+2011-07-04,15003.17,35,8450,Monday,7,2011
+2011-07-05,25992.260000000002,64,17634,Tuesday,7,2011
+2011-07-06,22960.48,66,17816,Wednesday,7,2011
+2011-07-07,27893.66,73,17654,Thursday,7,2011
+2011-07-08,17558.3,46,10165,Friday,7,2011
+2011-07-10,5993.87,24,4255,Sunday,7,2011
+2011-07-11,20080.03,48,14139,Monday,7,2011
+2011-07-12,16642.59,44,11029,Tuesday,7,2011
+2011-07-13,19432.85,56,15284,Wednesday,7,2011
+2011-07-14,30794.510000000002,69,16994,Thursday,7,2011
+2011-07-15,11857.300000000001,37,6379,Friday,7,2011
+2011-07-17,16958.6,50,10831,Sunday,7,2011
+2011-07-18,22018.36,46,12712,Monday,7,2011
+2011-07-19,46599.08,62,28409,Tuesday,7,2011
+2011-07-20,26086.97,52,15042,Wednesday,7,2011
+2011-07-21,29350.71,70,18885,Thursday,7,2011
+2011-07-22,14633.77,37,8333,Friday,7,2011
+2011-07-24,26796.920000000002,57,17578,Sunday,7,2011
+2011-07-25,19687.31,53,14276,Monday,7,2011
+2011-07-26,17293.001,46,12643,Tuesday,7,2011
+2011-07-27,20623.32,44,12756,Wednesday,7,2011
+2011-07-28,39094.69,78,26559,Thursday,7,2011
+2011-07-29,17240.61,62,11219,Friday,7,2011
+2011-07-31,26844.09,41,19271,Sunday,7,2011
+2011-08-01,19808.4,39,11308,Monday,8,2011
+2011-08-02,19027.05,41,13226,Tuesday,8,2011
+2011-08-03,26617.77,62,16147,Wednesday,8,2011
+2011-08-04,51621.97,79,34033,Thursday,8,2011
+2011-08-05,19825.4,52,12230,Friday,8,2011
+2011-08-07,7576.96,29,5185,Sunday,8,2011
+2011-08-08,19758.62,38,12442,Monday,8,2011
+2011-08-09,25057.12,40,15584,Tuesday,8,2011
+2011-08-10,19861.56,43,12333,Wednesday,8,2011
+2011-08-11,50482.57,63,35501,Thursday,8,2011
+2011-08-12,17970.170000000002,46,11049,Friday,8,2011
+2011-08-14,5718.57,25,3180,Sunday,8,2011
+2011-08-15,17243.97,48,10181,Monday,8,2011
+2011-08-16,16077.84,50,10474,Tuesday,8,2011
+2011-08-17,37616.26,51,21520,Wednesday,8,2011
+2011-08-18,51783.81,67,33641,Thursday,8,2011
+2011-08-19,17339.59,55,10671,Friday,8,2011
+2011-08-21,14566.84,38,8163,Sunday,8,2011
+2011-08-22,25891.18,62,14341,Monday,8,2011
+2011-08-23,22399.11,60,13707,Tuesday,8,2011
+2011-08-24,37291.01,72,27157,Wednesday,8,2011
+2011-08-25,22495.29,72,13175,Thursday,8,2011
+2011-08-26,23113.24,43,16308,Friday,8,2011
+2011-08-28,10805.03,37,6774,Sunday,8,2011
+2011-08-30,8833.710000000001,23,4215,Tuesday,8,2011
+2011-08-31,20540.84,42,11547,Wednesday,8,2011
+2011-09-01,37370.15,76,27857,Thursday,9,2011
+2011-09-02,26612.09,64,13877,Friday,9,2011
+2011-09-04,17005.03,49,10905,Sunday,9,2011
+2011-09-05,34810.7,67,21937,Monday,9,2011
+2011-09-06,25495.82,61,14471,Tuesday,9,2011
+2011-09-07,21967.420000000002,46,13731,Wednesday,9,2011
+2011-09-08,23188.3,72,16041,Thursday,9,2011
+2011-09-09,25142.15,58,16036,Friday,9,2011
+2011-09-11,35511.67,75,21146,Sunday,9,2011
+2011-09-12,27989.4,67,16438,Monday,9,2011
+2011-09-13,48162.25,63,35029,Tuesday,9,2011
+2011-09-14,22027.95,65,13438,Wednesday,9,2011
+2011-09-15,43854.57,77,21965,Thursday,9,2011
+2011-09-16,23248.98,46,14126,Friday,9,2011
+2011-09-18,15745.73,27,8994,Sunday,9,2011
+2011-09-19,45087.42,68,27472,Monday,9,2011
+2011-09-20,40861.91,63,20444,Tuesday,9,2011
+2011-09-21,37624.51,67,19632,Wednesday,9,2011
+2011-09-22,57869.36,111,32816,Thursday,9,2011
+2011-09-23,31781.100000000002,58,19800,Friday,9,2011
+2011-09-25,31372.661,75,19504,Sunday,9,2011
+2011-09-26,29329.841,65,15312,Monday,9,2011
+2011-09-27,29274.36,75,19533,Tuesday,9,2011
+2011-09-28,36767.89,80,23555,Wednesday,9,2011
+2011-09-29,44729.37,103,26974,Thursday,9,2011
+2011-09-30,36265.44,68,20064,Friday,9,2011
+2011-10-02,11582.95,34,8375,Sunday,10,2011
+2011-10-03,54053.0,62,23895,Monday,10,2011
+2011-10-04,35497.2,74,20978,Tuesday,10,2011
+2011-10-05,64121.43,94,44188,Wednesday,10,2011
+2011-10-06,53076.4,116,30407,Thursday,10,2011
+2011-10-07,40788.15,85,24218,Friday,10,2011
+2011-10-09,12466.81,38,7534,Sunday,10,2011
+2011-10-10,41700.32,90,24814,Monday,10,2011
+2011-10-11,41067.01,78,23189,Tuesday,10,2011
+2011-10-12,27731.37,82,16198,Wednesday,10,2011
+2011-10-13,33039.18,74,17897,Thursday,10,2011
+2011-10-14,32124.68,70,17215,Friday,10,2011
+2011-10-16,22010.96,35,8770,Sunday,10,2011
+2011-10-17,47005.24,85,30141,Monday,10,2011
+2011-10-18,37302.4,77,26900,Tuesday,10,2011
+2011-10-19,31265.46,76,18264,Wednesday,10,2011
+2011-10-20,59819.9,85,39169,Thursday,10,2011
+2011-10-21,38211.26,61,22304,Friday,10,2011
+2011-10-23,12339.16,42,7073,Sunday,10,2011
+2011-10-24,35887.840000000004,72,22633,Monday,10,2011
+2011-10-25,33523.95,72,24944,Tuesday,10,2011
+2011-10-26,30710.52,91,18769,Wednesday,10,2011
+2011-10-27,41515.020000000004,101,25355,Thursday,10,2011
+2011-10-28,34223.85,64,20120,Friday,10,2011
+2011-10-30,34571.23,96,20068,Sunday,10,2011
+2011-10-31,32146.99,63,15170,Monday,10,2011
+2011-11-01,29132.81,76,16577,Tuesday,11,2011
+2011-11-02,37774.7,82,23671,Wednesday,11,2011
+2011-11-03,45865.61,98,27170,Thursday,11,2011
+2011-11-04,54605.24,86,33248,Friday,11,2011
+2011-11-06,42941.340000000004,101,23305,Sunday,11,2011
+2011-11-07,28779.24,90,16439,Monday,11,2011
+2011-11-08,38295.12,99,21670,Tuesday,11,2011
+2011-11-09,57203.98,118,35122,Wednesday,11,2011
+2011-11-10,67815.13,124,37067,Thursday,11,2011
+2011-11-11,37081.37,93,23939,Friday,11,2011
+2011-11-13,28607.78,83,19764,Sunday,11,2011
+2011-11-14,56253.35,104,31846,Monday,11,2011
+2011-11-15,44627.48,104,25470,Tuesday,11,2011
+2011-11-16,48439.76,124,29156,Wednesday,11,2011
+2011-11-17,54760.3,136,30090,Thursday,11,2011
+2011-11-18,36751.25,103,20299,Friday,11,2011
+2011-11-20,30190.920000000002,98,18765,Sunday,11,2011
+2011-11-21,45333.13,96,24567,Monday,11,2011
+2011-11-22,46388.89,129,30635,Tuesday,11,2011
+2011-11-23,68279.87,130,37763,Wednesday,11,2011
+2011-11-24,38579.11,109,22923,Thursday,11,2011
+2011-11-25,25047.46,73,13736,Friday,11,2011
+2011-11-27,17300.96,56,10537,Sunday,11,2011
+2011-11-28,46714.91,114,26703,Monday,11,2011
+2011-11-29,43356.58,124,23087,Tuesday,11,2011
+2011-11-30,41481.23,99,24454,Wednesday,11,2011
+2011-12-01,44533.99,118,24857,Thursday,12,2011
+2011-12-02,40841.49,112,23011,Friday,12,2011
+2011-12-04,20375.96,62,11435,Sunday,12,2011
+2011-12-05,55647.45,116,37937,Monday,12,2011
+2011-12-06,43842.44,110,27459,Tuesday,12,2011
+2011-12-07,51918.71,100,34408,Wednesday,12,2011
+2011-12-08,39896.75,112,23745,Thursday,12,2011
+2011-12-09,15879.68,40,9587,Friday,12,2011

data/online_retail_cleaned.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:163b13e0c447277f6134967c6f1fc6b9bd9862e8028f3c5affbdd9e4ecb164df
+size 47685118

models/clv_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cedd59c79e59a346915e0930778475acbf92fce26e74634f5d3e2992ea86476
+size 10266689

models/kmeans_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a705465318e98090a96563bf874299fb9cc8fe60c9a74664c58e7e0e0b22484
+size 17975

models/scaler.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54b59ba462d0214c5673895b419bfba08f73ecfa9a3f98b668c17a4a8d124ec2
+size 927

requirements.txt

@@ -0,0 +1,8 @@
+fastapi==0.104.1
+uvicorn==0.24.0
+sqlalchemy==2.0.23
+psycopg2-binary==2.9.9
+pandas==2.2.0
+numpy==1.26.4
+joblib==1.3.2
+scikit-learn==1.8.0