Update app.py

app.py

@@ -58,7 +58,7 @@ HTTP_TRITON_URL = f"{TRITON_IP}:8000"
 TRITON_MODEL_NAME = "cloud-detection" # Ensure this matches your deployed model name
 TRITON_INPUT_NAME = "input_jp2_bytes" # Ensure this matches your model's config.pbtxt
 TRITON_OUTPUT_NAME = "output_mask" # Ensure this matches your model's config.pbtxt
-TRITON_TIMEOUT_SECONDS = 300.0 # 5 minutes timeout for connection/network
+TRITON_TIMEOUT_SECONDS = 300 # 5 minutes timeout for connection/network
 
 
 # --- Utility Functions (mostly from paste.txt) ---
@@ -403,7 +403,7 @@ def run_inference_triton_http(input_byte_array):
             inputs=inputs,
             outputs=outputs,
             request_id=str(os.getpid()), # Optional request ID
-            timeout=TRITON_TIMEOUT_SECONDS + 60.0 # Give extra time for the inference itself
+            timeout=TRITON_TIMEOUT_SECONDS
         )
         print("Triton inference request successful.")
         mask = response.as_numpy(TRITON_OUTPUT_NAME)