Redundant code?
Closed this issue · 2 comments
forresti commented
conv2_1_sep_bn = mx.symbol.BatchNorm(name='conv2_1_sep_bn', data=conv2_1_sep , use_global_stats=False, fix_gamma=False, eps=0.000100)
conv2_1_sep_scale = conv2_1_sep_bn
relu2_1_sep = mx.symbol.Activation(name='relu2_1_sep', data=conv2_1_sep_scale , act_type='relu')
Just curious -- what's the benefit of including the line conv2_1_sep_scale = conv2_1_sep_bn
?
In other words, would it be equivalent to have the following?
conv2_1_sep_bn = mx.symbol.BatchNorm(name='conv2_1_sep_bn', data=conv2_1_sep , use_global_stats=False, fix_gamma=False, eps=0.000100)
relu2_1_sep = mx.symbol.Activation(name='relu2_1_sep', data=conv2_1_sep_bn , act_type='relu')
KeyKy commented
conv2_1_sep_scale = conv2_1_sep_bn
is auto-generated by convert_symbol.py.
In my opinion, it should be equivalent to
conv2_1_sep_bn = mx.symbol.BatchNorm(name='conv2_1_sep_bn', data=conv2_1_sep , use_global_stats=False, fix_gamma=False, eps=0.000100)
relu2_1_sep = mx.symbol.Activation(name='relu2_1_sep', data=conv2_1_sep_bn , act_type='relu')
forresti commented
Thanks! I was guessing this was an artifact of some automated scripting, but I wanted to make sure I wasn't misunderstanding something.