Update app.py

app.py

@@ -1,18 +1,21 @@
-import fastapi
-from transformers 
-import pipeline
-import h5py
+import fastapifrom transformers import pipelineimport h5py
 
-# Create a context manager to load the model from the HDF5 file
-with h5py.File("model_finetuned.h5", "r") as f:
-    model = f["model"]
+# Load the model from the h5 file
+model = h5py.File("model_finetuned.h5", "r")["model"]
 
-# Define a dependency to inject the model into the `predict_endpoint()` function
-def get_model():
-    return model
+# Define a function to preprocess the image
+def preprocess_image(image):
+    # Resize the image to a fixed size
+    image = image.resize((224, 224))
+    # Convert the image to a NumPy array
+    image = np.array(image)
+    # Normalize the image
+    image = image / 255.0
+    # Return the image
+    return image
 
 # Define an endpoint to predict the output
-@app.post("/predict", dependencies=[Depends(get_model)])
+@app.post("/predict")
 async def predict_endpoint(image: fastapi.File):
     # Preprocess the image
     image = preprocess_image(image)
@@ -23,14 +26,6 @@ async def predict_endpoint(image: fastapi.File):
     # Return the prediction
     return {"prediction": prediction}
 
-# Define a FastAPI exception handler to handle any errors that occur while making predictions
-@app.exception_handler(Exception)
-async def handle_exception(request: Request, exc: Exception):
-    return JSONResponse(
-        status_code=exc.status_code,
-        content={"message": str(exc)},
-    )
-
 # Start the FastAPI app
 if __name__ == "__main__":
     import uvicorn