Potential bug in Hungarian matching interface

[tnybny]

Thanks for your contribution.

Aren't the arguments and parameters reversed for the call to merge here?

detr-tensorflow/detr_tf/loss/hungarian_matching.py

Line 182 in dda4c85

_p_bbox, _t_bbox = bbox.merge(p_bbox, t_bbox)

I say this because np_tf_linear_sum_assignment seems to expect the axis of the cost_matrix corresponding to the ground truth to be on the rows and corresponding to the predictions to be on the columns as per

detr-tensorflow/detr_tf/loss/hungarian_matching.py

Lines 30 to 31 in dda4c85

	target_indices = indices[0]
	pred_indices = indices[1]

whereas, the above call to merge would return tensors when operated upon to produce the a cost matrix where rows correspond to predictions and columns to ground truth.