About the pre-trained hypernetwork used for the results in Table 1, is it ofa_mbv3_d234_e346_k357_w1.0 or ofa_mbv3_d234_e346_k357_w1.2？

[yuny220]

Thank you very much for your outstanding work and sharing！May I ask if your results in the above figure(Table 1.) were obtained using the ofa_mbv3_d234_e346_k357_w1.0 or the ofa_mbv3_d234_e346_k357_w1.2 model, i.e. how big is the width multiplier？