Upload app.py with huggingface_hub

app.py

@@ -1,20 +1,18 @@
+import os
+import logging
+from typing import List, Dict
+from contextlib import asynccontextmanager
+from datetime import datetime, timezone
+
+from bs4 import BeautifulSoup
 from fastapi import FastAPI, Request, BackgroundTasks
-from fastapi.responses import HTMLResponse
-from fastapi.staticfiles import StaticFiles
 from fastapi.templating import Jinja2Templates
-import requests
-from bs4 import BeautifulSoup
+from fastapi.responses import JSONResponse
 import asyncio
 import aiohttp
-from datetime import datetime, timezone
-from typing import List, Dict, Optional
 import uvicorn
-import os
 import pandas as pd
 from datasets import Dataset, load_dataset
-from huggingface_hub import HfApi
-import logging
-from contextlib import asynccontextmanager
 
 # Configure logging
 logging.basicConfig(level=logging.INFO)
@@ -231,11 +229,14 @@ async def get_providers_data():
     # Sort by request count descending
     results.sort(key=lambda x: x["monthly_requests_int"], reverse=True)
     
-    return {
+    response = JSONResponse({
         "providers": results,
         "last_updated": datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
         "total_providers": len(results)
-    }
+    })
+    # Cache for 30 seconds to reduce load
+    response.headers["Cache-Control"] = "public, max-age=30"
+    return response
 
 @app.get("/api/providers/{provider}")
 async def get_provider_data(provider: str):
@@ -256,11 +257,13 @@ async def get_historical_data():
     """API endpoint to get historical data for line chart"""
     if not HF_TOKEN:
         logger.warning("No HF_TOKEN available for historical data")
-        return {
+        response = JSONResponse({
             "error": "Historical data not available - no HF token", 
             "historical_data": {},
             "message": "Historical data collection requires HuggingFace token"
-        }
+        })
+        response.headers["Cache-Control"] = "public, max-age=60"
+        return response
     
     try:
         # Load historical dataset
@@ -271,11 +274,13 @@ async def get_historical_data():
         
         if df.empty:
             logger.info("Dataset is empty - no historical data available yet")
-            return {
+            response = JSONResponse({
                 "historical_data": {},
                 "message": "No historical data available yet. Data collection is running - check back in 30 minutes.",
                 "last_updated": datetime.now().strftime('%Y-%m-%d %H:%M:%S')
-            }
+            })
+            response.headers["Cache-Control"] = "public, max-age=60"
+            return response
         
         # Group by timestamp and provider, get the latest entry for each timestamp-provider combo
         df['timestamp'] = pd.to_datetime(df['timestamp'])
@@ -324,14 +329,18 @@ async def get_historical_data():
         else:
             date_range = "No data"
 
-        return {
+        from fastapi.responses import JSONResponse
+        response = JSONResponse({
             "historical_data": historical_data,
             "last_updated": datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
             "total_data_points": total_data_points,
             "data_range": date_range,
             "earliest_date": df_filtered['timestamp'].min().isoformat() if not df_filtered.empty else None,
             "latest_date": df_filtered['timestamp'].max().isoformat() if not df_filtered.empty else None
-        }
+        })
+        # Cache for 2 minutes since historical data doesn't change as frequently
+        response.headers["Cache-Control"] = "public, max-age=120"
+        return response
         
     except Exception as e:
         logger.error(f"Error fetching historical data: {e}")
@@ -340,19 +349,23 @@ async def get_historical_data():
             logger.info("Dataset doesn't exist yet, triggering initial data collection")
             try:
                 await collect_and_store_data()
-                return {
+                response = JSONResponse({
                     "historical_data": {},
                     "message": "Dataset created! Historical data will appear after a few data collection cycles.",
                     "last_updated": datetime.now().strftime('%Y-%m-%d %H:%M:%S')
-                }
+                })
+                response.headers["Cache-Control"] = "public, max-age=60"
+                return response
             except Exception as create_error:
                 logger.error(f"Failed to create initial dataset: {create_error}")
         
-        return {
+        response = JSONResponse({
             "error": f"Failed to fetch historical data: {str(e)}", 
             "historical_data": {},
             "message": "Historical data temporarily unavailable"
-        }
+        })
+        response.headers["Cache-Control"] = "public, max-age=30"
+        return response
 
 @app.get("/api/models")
 async def get_provider_models_data():
@@ -400,14 +413,17 @@ async def get_provider_models_data():
     for provider in PROVIDERS:
         provider_totals[provider] = len(provider_models.get(provider, set()))
     
-    return {
+    response = JSONResponse({
         "matrix": matrix,
         "providers": PROVIDERS,
         "provider_totals": provider_totals,
         "provider_mapping": PROVIDER_TO_INFERENCE_NAME,
         "total_models": len(all_models),
         "last_updated": datetime.now().strftime('%Y-%m-%d %H:%M:%S')
-    }
+    })
+    # Cache for 5 minutes since models data changes infrequently
+    response.headers["Cache-Control"] = "public, max-age=300"
+    return response
 
 @app.post("/api/collect-now")
 async def trigger_data_collection(background_tasks: BackgroundTasks):