engine = builder.build_cuda_engine(network) 之后如何导出engine模型

After building the TensorRT engine with `builder.build_cuda_engine(network)`, you can save the engine to a file using the `serialize()` method of the engine object. Here is an example:

import tensorrt as trt

# Create a builder object and build the engine
builder = trt.Builder(trt.Logger(trt.Logger.INFO))
engine = builder.build_cuda_engine(network)

# Save the engine to a file
with open("engine.trt", "wb") as f:
    f.write(engine.serialize())

In this example, we create a builder object and build the engine using a network object. We then save the engine to a file named "engine.trt" using the `serialize()` method of the engine object. This will write the engine to a binary file that can be loaded later using the `deserialize_cuda_engine()` method of the TensorRT runtime object.