Upload folder using huggingface_hub

Dockerfile

@@ -1,14 +1,14 @@
-# Dockerfile — FastAPI backend for HF Spaces (Docker SDK)
+# FastAPI backend for HF Spaces
 FROM python:3.10-slim
 
-# (Optional) system deps helpful for pandas/numpy wheels
+# Optional build deps for numpy/pandas
 RUN apt-get update && apt-get install -y --no-install-recommends \
     build-essential gcc g++ && \
     rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 
-# Install Python dependencies first (better layer caching)
+# Install Python dependencies
 COPY requirements.txt /app/requirements.txt
 RUN pip install --no-cache-dir --upgrade pip && \
     pip install --no-cache-dir -r requirements.txt
@@ -16,3 +16,10 @@ RUN pip install --no-cache-dir --upgrade pip && \
 # Copy app + model
 COPY app.py /app/app.py
 COPY SuperKart_prediction_model_v1_0.joblib /app/SuperKart_prediction_model_v1_0.joblib
+
+# Hugging Face Spaces expects PORT
+ENV PORT=7860
+EXPOSE 7860
+
+# Start with uvicorn
+CMD ["bash", "-lc", "uvicorn app:app --host 0.0.0.0 --port ${PORT}"]

app.py

@@ -6,14 +6,13 @@ from fastapi import FastAPI, File, UploadFile, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel
 
-MODEL_PATH = "SuperKart_prediction_model_v1_0.joblib"  # <-- update if your filename differs
+MODEL_PATH = "SuperKart_prediction_model_v1_0.joblib"
 
-# Expected feature order/names (must match training)
 EXPECTED_COLS = [
     "Product_Weight",
     "Product_Allocated_Area",
     "Product_MRP",
-    "Store_Age",  # change to Store_Age_Years or Store_Establishment_Year if that's what you trained on
+    "Store_Age",
     "Product_Sugar_Content",
     "Product_Type",
     "Store_Type",
@@ -21,45 +20,40 @@ EXPECTED_COLS = [
     "Store_Location_City_Type",
 ]
 
-# ---------- App ----------
 app = FastAPI(title="SuperKart Backend", version="1.0")
 
-# CORS (allow any frontend, incl. your Streamlit Space)
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],  # tighten for prod
+    allow_origins=["*"],
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
 )
 
-# ---------- Load model at startup ----------
 model = None
+
 @app.on_event("startup")
 def load_model():
     global model
     model = joblib.load(MODEL_PATH)
 
-# ---------- Schemas ----------
 class Payload(BaseModel):
     Product_Weight: float
     Product_Allocated_Area: float
     Product_MRP: float
-    Store_Age: int         # adjust type/name if needed
+    Store_Age: int
     Product_Sugar_Content: str
     Product_Type: str
     Store_Type: str
-    Store_Size: int        # or str, if you trained as categorical strings
-    Store_Location_City_Type: int  # or str, adjust to your training
+    Store_Size: int
+    Store_Location_City_Type: int
 
-# ---------- Helpers ----------
 def validate_and_order(df: pd.DataFrame) -> pd.DataFrame:
     missing = [c for c in EXPECTED_COLS if c not in df.columns]
     if missing:
         raise HTTPException(status_code=422, detail=f"Missing columns: {missing}")
     return df[EXPECTED_COLS].copy()
 
-# ---------- Endpoints ----------
 @app.get("/health")
 def health():
     return {"status": "ok"}
@@ -77,16 +71,14 @@ def predict_single(payload: Payload):
 @app.post("/v1/salespricebatch")
 def predict_batch(file: UploadFile = File(...)):
     try:
-        # read CSV into DataFrame
         content = file.file.read()
         df = pd.read_csv(io.BytesIO(content))
         X = validate_and_order(df)
         y = model.predict(X)
-        out = df.copy()
-        out["Predicted Price"] = y
-        # return as records (JSON)
-        return json.loads(out.to_json(orient="records"))
+        df["Predicted Price"] = y
+        return json.loads(df.to_json(orient="records"))
     except HTTPException:
         raise
     except Exception as e:
         raise HTTPException(status_code=500, detail=str(e))
+

requirements.txt

@@ -5,4 +5,3 @@ numpy==2.0.2
 scikit-learn==1.6.1
 xgboost==2.1.4
 joblib==1.4.2
-python-multipart==0.0.9