关于Export onnx的问题

[peterchen183]

        # Export the model
        torch.onnx.export(self.net,
                        im,
                        f               = model_path,
                        verbose         = False,
                        opset_version   = 12,
                        training        = torch.onnx.TrainingMode.EVAL,
                        do_constant_folding = True,
                        input_names     = input_layer_names,
                        output_names    = output_layer_names,
                        dynamic_axes    = None)

请问支持dynamic_axes不是None的情况吗?如果支持，请问该如何修改，谢谢

[bubbliiiing]

dynamic_axes是干嘛的……0 0

[peterchen183]

dynamic_axes是干嘛的……0 0

        torch.onnx.export(self.net,
                        im,
                        f               = model_path,
                        verbose         = False,
                        opset_version   = 12,
                        training        = torch.onnx.TrainingMode.EVAL,
                        do_constant_folding = True,
                        input_names     = input_layer_names,
                        output_names    = output_layer_names,
                        dynamic_axes    ={
                                            'images': {
                                                0: 'batch'},
                                            'output': {
                                                0: 'batch'}
                                        })

onnx转engine(trt)的时候，支持动态输入（Dynamic shapes），这样可以做batch inference，提高一点儿效率

[bubbliiiing]

呀？那……不设置的话是不是默认为1

[peterchen183]

呀？那……不设置的话是不是默认为1

应该算是吧 只能single inference了 相当于batch inference但batch_size为1