Spaces:
Runtime error
Runtime error
Commit ·
c82f84d
1
Parent(s): aee0892
Upload E-Commerce Product Intelligence Dashboard (frontend + backend)
Browse files- backend/app.py +12 -35
backend/app.py
CHANGED
|
@@ -13,16 +13,13 @@ logger = logging.getLogger(__name__)
|
|
| 13 |
|
| 14 |
app = FastAPI(title="E-Commerce Product Intelligence Platform")
|
| 15 |
|
| 16 |
-
# HF Dataset config
|
| 17 |
HF_DATASET_ID = "Vincentran/ecommerce-dataset"
|
| 18 |
HF_CSV_PATH = "data/ecommerce_products.csv"
|
| 19 |
|
| 20 |
-
# Cache DataFrame
|
| 21 |
_data_cache = None
|
| 22 |
|
| 23 |
|
| 24 |
def load_data():
|
| 25 |
-
"""Load CSV từ HF Dataset with cache."""
|
| 26 |
try:
|
| 27 |
if _data_cache is not None:
|
| 28 |
logger.info("Using cached DataFrame")
|
|
@@ -44,7 +41,6 @@ def load_data():
|
|
| 44 |
df = pd.read_csv(local_csv_path)
|
| 45 |
logger.info(f"Loaded {len(df)} rows, columns: {list(df.columns)}")
|
| 46 |
|
| 47 |
-
# Cache DataFrame
|
| 48 |
_data_cache = df
|
| 49 |
return df
|
| 50 |
|
|
@@ -54,7 +50,6 @@ def load_data():
|
|
| 54 |
|
| 55 |
|
| 56 |
def refresh_cache():
|
| 57 |
-
"""Refresh data cache."""
|
| 58 |
_data_cache = None
|
| 59 |
return load_data()
|
| 60 |
|
|
@@ -65,10 +60,7 @@ def root():
|
|
| 65 |
|
| 66 |
|
| 67 |
@app.get("/data")
|
| 68 |
-
def get_data(
|
| 69 |
-
page: int = Query(1, ge=1, description="Page number"),
|
| 70 |
-
limit: int = Query(100, ge=1, le=500, description="Items per page")
|
| 71 |
-
):
|
| 72 |
df = load_data()
|
| 73 |
total = len(df)
|
| 74 |
start = (page - 1) * limit
|
|
@@ -123,7 +115,6 @@ def stats_rating():
|
|
| 123 |
|
| 124 |
@app.get("/stats/price-range")
|
| 125 |
def stats_price_range():
|
| 126 |
-
"""Price distribution by range."""
|
| 127 |
df = load_data()
|
| 128 |
if "price" not in df.columns:
|
| 129 |
raise HTTPException(status_code=400, detail="Missing 'price' column")
|
|
@@ -153,15 +144,10 @@ def insights():
|
|
| 153 |
|
| 154 |
|
| 155 |
@app.get("/search")
|
| 156 |
-
def search(
|
| 157 |
-
query: str = Query(..., description="Search query"),
|
| 158 |
-
page: int = Query(1, ge=1, description="Page number"),
|
| 159 |
-
limit: int = Query(100, ge=1, le=500, description="Items per page")
|
| 160 |
-
):
|
| 161 |
df = load_data()
|
| 162 |
q = query.lower()
|
| 163 |
|
| 164 |
-
# Search only in important columns
|
| 165 |
search_cols = ["product_name", "category", "brand", "description"]
|
| 166 |
search_cols = [col for col in search_cols if col in df.columns]
|
| 167 |
|
|
@@ -192,16 +178,15 @@ def search(
|
|
| 192 |
|
| 193 |
@app.get("/filter")
|
| 194 |
def filter_products(
|
| 195 |
-
category: Optional[str] = Query(None
|
| 196 |
-
min_price: Optional[float] = Query(None
|
| 197 |
-
max_price: Optional[float] = Query(None
|
| 198 |
-
min_rating: Optional[float] = Query(None
|
| 199 |
-
page: int = Query(1, ge=1
|
| 200 |
-
limit: int = Query(100, ge=1, le=500
|
| 201 |
):
|
| 202 |
df = load_data()
|
| 203 |
|
| 204 |
-
# Apply filters
|
| 205 |
if category and "category" in df.columns:
|
| 206 |
df = df[df["category"] == category]
|
| 207 |
if min_price and "price" in df.columns:
|
|
@@ -221,12 +206,7 @@ def filter_products(
|
|
| 221 |
data = df.iloc[start:end].to_dict("records")
|
| 222 |
return {
|
| 223 |
"data": data,
|
| 224 |
-
"filters": {
|
| 225 |
-
"category": category,
|
| 226 |
-
"min_price": min_price,
|
| 227 |
-
"max_price": max_price,
|
| 228 |
-
"min_rating": min_rating
|
| 229 |
-
},
|
| 230 |
"page": page,
|
| 231 |
"limit": limit,
|
| 232 |
"total": total,
|
|
@@ -235,7 +215,7 @@ def filter_products(
|
|
| 235 |
|
| 236 |
|
| 237 |
@app.get("/recommend")
|
| 238 |
-
def recommend(category: str, limit: int = Query(10, ge=1, le=50
|
| 239 |
df = load_data()
|
| 240 |
if "category" not in df.columns:
|
| 241 |
raise HTTPException(status_code=400, detail="Missing 'category' column")
|
|
@@ -252,7 +232,6 @@ def recommend(category: str, limit: int = Query(10, ge=1, le=50, description="Nu
|
|
| 252 |
|
| 253 |
@app.post("/refresh-data")
|
| 254 |
def refresh_data():
|
| 255 |
-
"""Refresh data cache from HF Dataset."""
|
| 256 |
try:
|
| 257 |
df = refresh_cache()
|
| 258 |
return {"status": "Data refreshed successfully", "rows": len(df)}
|
|
@@ -262,21 +241,19 @@ def refresh_data():
|
|
| 262 |
|
| 263 |
@app.post("/run-scraper")
|
| 264 |
def trigger_scraper():
|
| 265 |
-
"""Trigger download Kaggle → save CSV → upload to HF."""
|
| 266 |
import subprocess
|
| 267 |
result = subprocess.run(["python", "backend/scraper.py"], capture_output=True, text=True)
|
| 268 |
if result.returncode == 0:
|
| 269 |
-
# Refresh cache after scraper
|
| 270 |
refresh_cache()
|
| 271 |
return {"status": "Scraper completed successfully", "output": result.stdout}
|
| 272 |
else:
|
| 273 |
return {"status": "Scraper failed", "error": result.stderr}
|
| 274 |
|
| 275 |
|
| 276 |
-
# Mount frontend
|
| 277 |
frontend_dir = Path("frontend")
|
| 278 |
if frontend_dir.exists():
|
| 279 |
-
app.mount("/", StaticFiles(directory=str(frontend_dir), html=True), name="frontend")
|
| 280 |
else:
|
| 281 |
@app.get("/")
|
| 282 |
def frontend_placeholder():
|
|
|
|
| 13 |
|
| 14 |
app = FastAPI(title="E-Commerce Product Intelligence Platform")
|
| 15 |
|
|
|
|
| 16 |
HF_DATASET_ID = "Vincentran/ecommerce-dataset"
|
| 17 |
HF_CSV_PATH = "data/ecommerce_products.csv"
|
| 18 |
|
|
|
|
| 19 |
_data_cache = None
|
| 20 |
|
| 21 |
|
| 22 |
def load_data():
|
|
|
|
| 23 |
try:
|
| 24 |
if _data_cache is not None:
|
| 25 |
logger.info("Using cached DataFrame")
|
|
|
|
| 41 |
df = pd.read_csv(local_csv_path)
|
| 42 |
logger.info(f"Loaded {len(df)} rows, columns: {list(df.columns)}")
|
| 43 |
|
|
|
|
| 44 |
_data_cache = df
|
| 45 |
return df
|
| 46 |
|
|
|
|
| 50 |
|
| 51 |
|
| 52 |
def refresh_cache():
|
|
|
|
| 53 |
_data_cache = None
|
| 54 |
return load_data()
|
| 55 |
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
@app.get("/data")
|
| 63 |
+
def get_data(page: int = Query(1, ge=1), limit: int = Query(100, ge=1, le=500)):
|
|
|
|
|
|
|
|
|
|
| 64 |
df = load_data()
|
| 65 |
total = len(df)
|
| 66 |
start = (page - 1) * limit
|
|
|
|
| 115 |
|
| 116 |
@app.get("/stats/price-range")
|
| 117 |
def stats_price_range():
|
|
|
|
| 118 |
df = load_data()
|
| 119 |
if "price" not in df.columns:
|
| 120 |
raise HTTPException(status_code=400, detail="Missing 'price' column")
|
|
|
|
| 144 |
|
| 145 |
|
| 146 |
@app.get("/search")
|
| 147 |
+
def search(query: str = Query(...), page: int = Query(1, ge=1), limit: int = Query(100, ge=1, le=500)):
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
df = load_data()
|
| 149 |
q = query.lower()
|
| 150 |
|
|
|
|
| 151 |
search_cols = ["product_name", "category", "brand", "description"]
|
| 152 |
search_cols = [col for col in search_cols if col in df.columns]
|
| 153 |
|
|
|
|
| 178 |
|
| 179 |
@app.get("/filter")
|
| 180 |
def filter_products(
|
| 181 |
+
category: Optional[str] = Query(None),
|
| 182 |
+
min_price: Optional[float] = Query(None),
|
| 183 |
+
max_price: Optional[float] = Query(None),
|
| 184 |
+
min_rating: Optional[float] = Query(None),
|
| 185 |
+
page: int = Query(1, ge=1),
|
| 186 |
+
limit: int = Query(100, ge=1, le=500)
|
| 187 |
):
|
| 188 |
df = load_data()
|
| 189 |
|
|
|
|
| 190 |
if category and "category" in df.columns:
|
| 191 |
df = df[df["category"] == category]
|
| 192 |
if min_price and "price" in df.columns:
|
|
|
|
| 206 |
data = df.iloc[start:end].to_dict("records")
|
| 207 |
return {
|
| 208 |
"data": data,
|
| 209 |
+
"filters": {"category": category, "min_price": min_price, "max_price": max_price, "min_rating": min_rating},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 210 |
"page": page,
|
| 211 |
"limit": limit,
|
| 212 |
"total": total,
|
|
|
|
| 215 |
|
| 216 |
|
| 217 |
@app.get("/recommend")
|
| 218 |
+
def recommend(category: str, limit: int = Query(10, ge=1, le=50)):
|
| 219 |
df = load_data()
|
| 220 |
if "category" not in df.columns:
|
| 221 |
raise HTTPException(status_code=400, detail="Missing 'category' column")
|
|
|
|
| 232 |
|
| 233 |
@app.post("/refresh-data")
|
| 234 |
def refresh_data():
|
|
|
|
| 235 |
try:
|
| 236 |
df = refresh_cache()
|
| 237 |
return {"status": "Data refreshed successfully", "rows": len(df)}
|
|
|
|
| 241 |
|
| 242 |
@app.post("/run-scraper")
|
| 243 |
def trigger_scraper():
|
|
|
|
| 244 |
import subprocess
|
| 245 |
result = subprocess.run(["python", "backend/scraper.py"], capture_output=True, text=True)
|
| 246 |
if result.returncode == 0:
|
|
|
|
| 247 |
refresh_cache()
|
| 248 |
return {"status": "Scraper completed successfully", "output": result.stdout}
|
| 249 |
else:
|
| 250 |
return {"status": "Scraper failed", "error": result.stderr}
|
| 251 |
|
| 252 |
|
| 253 |
+
# ✅ Mount frontend at /frontend (not /)
|
| 254 |
frontend_dir = Path("frontend")
|
| 255 |
if frontend_dir.exists():
|
| 256 |
+
app.mount("/frontend", StaticFiles(directory=str(frontend_dir), html=True), name="frontend")
|
| 257 |
else:
|
| 258 |
@app.get("/")
|
| 259 |
def frontend_placeholder():
|