onnx decoder batch foward

[kyz-xxl]

I would like to know how the input format for the decoder part is when inferring in ONNX format; is the dynamic dimension set like this?
dynamic_axes = {"image_embeddings": {0: "batch_size"},
"point_coords": {0: 'batch_size', 1: "num_points"},
"point_labels": {0: 'batch_size', 1: "num_points"},
"mask_input": {0: "batch_size"},
"has_mask_input": {0: "batch_size"},
"masks": {0: "batch_size"},
"iou_predictions": {0: "batch_size"},
"low_res_masks": {0: "batch_size"}
}
If not, how should it be modified?