error changes

app.py

@@ -4,16 +4,29 @@ from pydantic import BaseModel
 from tensorflow.keras.models import load_model
 from tensorflow.keras.preprocessing.text import tokenizer_from_json
 from tensorflow.keras.preprocessing.sequence import pad_sequences
+import numpy as np
 import json
 from typing import Union, List
 
 app = FastAPI()
 
-# Load model and tokenizer
-model = load_model('news_classifier.h5')
-with open('tokenizer.json', 'r') as f:
-    tokenizer_data = json.load(f)
-    tokenizer = tokenizer_from_json(tokenizer_data)
+# Global variables for model and tokenizer
+model = None
+tokenizer = None
+
+def load_model_and_tokenizer():
+    global model, tokenizer
+    try:
+        model = load_model('news_classifier.h5')
+        with open('tokenizer.json', 'r') as f:
+            tokenizer_data = json.load(f)
+            tokenizer = tokenizer_from_json(tokenizer_data)
+    except Exception as e:
+        print(f"Error loading model or tokenizer: {str(e)}")
+        raise e
+
+# Load on startup
+load_model_and_tokenizer()
 
 class PredictionInput(BaseModel):
     text: Union[str, List[str]]
@@ -22,35 +35,54 @@ class PredictionOutput(BaseModel):
     label: str
     score: float
 
-@app.post("/predict")
+@app.get("/")
+def read_root():
+    return {
+        "message": "News Source Classifier API",
+        "model_type": "LSTM",
+        "version": "1.0",
+        "status": "ready" if model and tokenizer else "not_loaded"
+    }
+
+@app.post("/predict", response_model=Union[PredictionOutput, List[PredictionOutput]])
 async def predict(input_data: PredictionInput):
+    if not model or not tokenizer:
+        try:
+            load_model_and_tokenizer()
+        except Exception as e:
+            raise HTTPException(status_code=500, detail="Model not loaded")
+    
     try:
-        # Convert input to list if it's a single string
+        # Handle both single string and list inputs
         texts = input_data.text if isinstance(input_data.text, list) else [input_data.text]
         
         # Preprocess
         sequences = tokenizer.texts_to_sequences(texts)
-        padded = pad_sequences(sequences, maxlen=41)  # Use your model's expected input length
+        padded = pad_sequences(sequences, maxlen=41)  # Match your model's input length
         
-        # Predict
-        predictions = model.predict(padded)
+        # Get predictions
+        predictions = model.predict(padded, verbose=0)
         
-        # Format results
+        # Process results
         results = []
         for pred in predictions:
-            score = float(pred[1])  # Assuming binary classification
-            label = "foxnews" if score > 0.5 else "nbc"
+            label = "foxnews" if pred[1] > 0.5 else "nbc"
+            score = float(pred[1] if label == "foxnews" else 1 - pred[1])
             results.append({
                 "label": label,
-                "score": score if label == "foxnews" else 1 - score
+                "score": score
             })
         
         # Return single result if input was single string
         return results[0] if isinstance(input_data.text, str) else results
-            
+    
     except Exception as e:
         raise HTTPException(status_code=500, detail=str(e))
 
-@app.get("/")
-async def root():
-    return {"message": "News Classifier API is running"}
+@app.post("/reload")
+async def reload_model():
+    try:
+        load_model_and_tokenizer()
+        return {"message": "Model reloaded successfully"}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

requirements.txt

@@ -3,5 +3,4 @@ fastapi>=0.68.0
 uvicorn>=0.15.0
 pydantic>=1.8.2
 numpy>=1.19.2
-scikit-learn>=0.24.2
 python-multipart