Upload app.py

app.py

@@ -1,25 +1,42 @@
 # app.py
-from transformers import pipeline
 from fastapi import FastAPI, HTTPException
 import torch
+from torch import nn
+from torch.utils.data import Dataset, DataLoader
+import pandas as pd
 from typing import List, Dict
-import json
+from train import MangaRecommender, MangaDataset  # Import from train.py
 
 app = FastAPI()
 
-# Load model and mappings
-checkpoint = torch.load('manga_recommender.pt')
-model = MangaRecommender(
-    num_users=len(checkpoint['user_mapping']),
-    num_items=len(checkpoint['manga_mapping'])
-)
-model.load_state_dict(checkpoint['model_state_dict'])
-user_mapping = checkpoint['user_mapping']
-manga_mapping = checkpoint['manga_mapping']
-reverse_manga_mapping = {v: k for k, v in manga_mapping.items()}
+try:
+    # Load model and mappings
+    checkpoint = torch.load('manga_recommender.pt')
+    model = MangaRecommender(
+        num_users=len(checkpoint['user_mapping']),
+        num_items=len(checkpoint['manga_mapping'])
+    )
+    model.load_state_dict(checkpoint['model_state_dict'])
+    user_mapping = checkpoint['user_mapping']
+    manga_mapping = checkpoint['manga_mapping']
+    reverse_manga_mapping = {v: k for k, v in manga_mapping.items()}
+    print("Model loaded successfully")
+except Exception as e:
+    print(f"Error loading model: {e}")
+    model = None
+    user_mapping = {}
+    manga_mapping = {}
+    reverse_manga_mapping = {}
+
+@app.get("/")
+async def root():
+    return {"status": "running", "model_loaded": model is not None}
 
 @app.post("/predict")
 async def predict(user_id: str, top_k: int = 10):
+    if model is None:
+        raise HTTPException(status_code=500, detail="Model not loaded")
+        
     try:
         # Get user index
         user_idx = user_mapping.get(user_id)
@@ -43,10 +60,16 @@ async def predict(user_id: str, top_k: int = 10):
                 "scores": scores
             }
     except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
+        raise HTTPException(
+            status_code=500, 
+            detail=f"Prediction error: {str(e)}"
+        )
 
 @app.post("/update")
 async def update_model(ratings: List[Dict]):
+    if model is None:
+        raise HTTPException(status_code=500, detail="Model not loaded")
+        
     try:
         # Convert ratings to training format
         df = pd.DataFrame(ratings)
@@ -62,12 +85,14 @@ async def update_model(ratings: List[Dict]):
         criterion = nn.MSELoss()
         
         model.train()
+        total_loss = 0
         for user, item, rating in loader:
             optimizer.zero_grad()
             pred = model(user, item)
             loss = criterion(pred, rating)
             loss.backward()
             optimizer.step()
+            total_loss += loss.item()
             
         # Save updated model
         torch.save({
@@ -76,6 +101,28 @@ async def update_model(ratings: List[Dict]):
             'manga_mapping': manga_mapping
         }, 'manga_recommender.pt')
         
-        return {"message": "Model updated successfully"}
+        return {
+            "message": "Model updated successfully",
+            "average_loss": total_loss / len(loader)
+        }
     except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
+        raise HTTPException(
+            status_code=500, 
+            detail=f"Update error: {str(e)}"
+        )
+
+@app.get("/model-info")
+async def model_info():
+    if model is None:
+        raise HTTPException(status_code=500, detail="Model not loaded")
+        
+    return {
+        "num_users": len(user_mapping),
+        "num_manga": len(manga_mapping),
+        "embedding_size": model.user_factors.embedding_dim
+    }
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)
+    raise HTTPException(status_code=500, detail=str(e))