debug: fix the first batch output unsync issue

samples/SampleTensorRTResnet.py

@@ -1169,6 +1169,7 @@ def run_inference(self, tensor_image) -> str:
         for out in self.outputs:
             cuda.memcpy_dtoh_async(out.host, out.device, self.stream)
 
+        self.stream.synchronize()
         # Find most probable image type and return resnet categoy description
         [result] = [out.host for out in self.outputs]
         return resnet_categories[np.argmax(result)]
@@ -1311,7 +1312,7 @@ def infer_on_video(gpu_id: int, input_video: str, trt_nn_file: str):
 if __name__ == "__main__":
     if len(sys.argv) < 3:
         print("Provide gpu id and path to input video file.")
-        exit
+        exit(1)
 
     gpu_id = int(sys.argv[1])
     input_video = sys.argv[2]

setup.py

@@ -38,9 +38,12 @@ def get_cupy() -> str:
             "dev": ["pycuda", "pyopengl", "torch", "torchvision", "opencv-python", "onnx", "tensorrt", f"PytorchNvCodec @ file://{os.getcwd()}/src/PytorchNvCodec/"],
             "samples": ["pycuda", "pyopengl", "torch", "torchvision", "opencv-python", "onnx", "tensorrt", "tqdm", cupy, PytorchNvCodec],
             "tests": ["pycuda", "pyopengl", "torch", "torchvision", "opencv-python", PytorchNvCodec],
-            "torch": ["torch", "torchvision", "opencv-python", PytorchNvCodec],
+            "torch": ["torch", "torchvision", PytorchNvCodec],
             "tensorrt": ["torch", "torchvision", cupy, PytorchNvCodec],
         },
+        dependency_links=[
+            "https://pypi.ngc.nvidia.com"
+        ],
         packages=["PyNvCodec"],
         package_data={"PyNvCodec": ["__init__.pyi"]},
         package_dir={"": "src"},