The example cannot work when plugin to a large model

[hikaru-nara]

When I use the method in this repo to export F.grid_sample in a large model, there is something wrong with the exported graph. The shapes of the inputs of GridSampler nodes are nonetype.

To reproduce

Change the model definition to (Note the +1 in the argument passed to F.grid_sample)

class MyModel(torch.nn.Module):
    def __init__(self):
        super(MyModel,self).__init__()

    def forward(self, input, grid):
        return F.grid_sample(input+1, grid, mode='bilinear', padding_mode='zeros', align_corners=True)

Then export the model, and print the nodeof op GridSampler with the following code.

def check_onnx(onnx_model_file):
    graph = gs.import_onnx(onnx.load(onnx_model_file))
    assert(graph is not None)
    for node in graph.nodes:
        if node.op == 'GridSampler':
            print(node)

And then the console prints

GridSampler_2 (GridSampler)
        Inputs: [Variable (3): (shape=None, dtype=None), Variable (grid): (shape=[1, 1, 128000, 2], dtype=float16)]
        Outputs: [Variable (4): (shape=[1, 17, 1, 128000], dtype=float16)]
Attributes: OrderedDict([('aligncorners', 1), ('interpolationmode', 0), ('paddingmode', 0)])

The desired behavior is that the shape and dtype of the input should not be None.

[hikaru-nara]

It is weird that adding that plus one cause the method to fail.... Am I doing anything wrong? Please, any hint will be appreciated.

[hikaru-nara]

It seems that it is ok even though the shapes are none. I've not tested but the engine has been built successfully. Close this one for the time being.

[hikaru-nara]

You just need to manually provide the shapes for modify_onnx().

[jucic]

You just need to manually provide the shapes for modify_onnx().

I encountered with the same problem, however, gridsampler is used many times with different input shape, so do you know any other solution for this issue, thanks ahead!