RunTimeError for dimension specs

[harrismathew]

I'm getting the following error when trying to compile my model with fp32 inputs.

RuntimeError: [Error thrown at core/ir/ir.cpp:41] Expected vals.size() == specs.size() to be true but got false
Expected dimension specifications for all input tensors, but found 2 input tensors and 1 dimension specs

What does it mean? My model requires two inputs of the same shape and size. My forward method simply looks something like:

    ...
    def forward(self, inp_1, inp_2):
        return self.path_1(inp_1) + self.path_2(inp_2)

[ksivaman]

Even if the sizes are same you'll still have to specify shapes for both inputs, I believe something like this would work

inp_d = torch_tensorrt.Input(shape=<INSERT_SHAPE>, dtype=torch_tensorrt.dtype.float32)
module_ts = torch_tensorrt.compile(model, inputs=[inp_d, inp_d])