I’m trying to convert an ONNX model, built from Pytorch, into TensorRT 22.214.171.124.
Unfortunately, it can’t handle a “Greater” layer.
How can I solve this issue? is it possible to implement this layer myself, when loading from ONNX? Is there another recommended way to convert a Pytorch model to TensorRT in which a straightforward layer such as “Greater” has an implementation, or a way to implement it myself?