-
Notifications
You must be signed in to change notification settings - Fork 32
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
RowVector support in v0.7 #531
Comments
I would suggest to remove the output type annotation altogether. In v0.7 you will get a |
PS: there is no performance impact on this change, and the function is type stable after inspecting with |
Okay, since there seems to be no good solution, I agree. We should add a note in the comment, though, and maybe add a unit test to check that type in the respective version. |
👍 for adding a comment in the docstring |
#531 - Remove result type from SparseMatrixExp's 'get_row' function
This is the last missing step toward v0.7 support (#121).
In
ExponentialMap.jl
we have a function with this signature:We currently use
transpose
, which in v0.7 results in an object of typeLinearAlgebra.Transpose
, so we get an error because the type is different from the expectedRowVector
. In v0.6 thetranspose
function creates aRowVector
.I cannot find a way to create a
RowVector
without deprecation warning in v0.7.To me, the suggested way to transpose a vector is either through
transpose
(lazy) orpermutedims
(explicit) (search for "transpose" here). However, the types are different fromRowVector
:The above is the only place where we use
RowVector
, so it might be safe to just change the return type of this function. Should we do that? To which type?Note that we need a solution that also works in v0.6, but we can always write a helper function that does the right thing.
We can also create a
Vector
out of it:The text was updated successfully, but these errors were encountered: