You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Note, in principle, DotLayer would not need any options anymore, as we can infer all the information about what axes to reduce and what axes are shared. However, I think it is good to have to specify the reduce axes explicitly. Maybe we could however omit this when the user specifies out_shape.
The reduce arg is always needed to distinguish it from the common axes, right?
(Okay, you could infer it from out_shape as it is not present there but the common axes are.)
The reduce arg is always needed to distinguish it from the common axes, right? (Okay, you could infer it from out_shape as it is not present there but the common axes are.)
This issue is just to collect recent orthogonal but related proposals.
DotLayer
specificDotLayer
defaults should not use integers, change defaults #627DotLayer
, use singlereduce
argument #636DotLayer
could infervar1
/var2
automatically #738DotLayer
: maybe use 1D conv CuDNN kernel #740DotLayer
to allow sparse inputs #741Related on dim tags
Other related
GatherLayer
LinearLayer
The text was updated successfully, but these errors were encountered: