From a code perspective, sparse convolution operations only appear in test_ forward. Does this mean that the model only trains traditional classification and regression heads? And the process shown in Figure 1 in the paper only appears during the testing phase?
xuwanzhuang
commented
5 months ago
From a code perspective, sparse convolution operations only appear in test_ forward. Does this mean that the model only trains traditional classification and regression heads? And the process shown in Figure 1 in the paper only appears during the testing phase?