Loading…
torch2trt is a PyTorch to TensorRT converter that uses the TensorRT Python API to optimize models. It is easy to use—convert modules with a single function call (torch2trt)—and easy to extend by writing your own layer converters registered with @tensorrt_converter. The project notes that TensorRT/PyTorch coverage is limited and it was created primarily to optimize models used in the JetBot project, with feedback encouraged.
We haven’t found torch2trt in the AI answers we’ve checked.