minimal set of fixes to get tests passing in julia 0.5 #18
Conversation
|
The package will pass all tests with a more minimalist set of changes:
As per the
|
|
To get it through Travis you may want to remove the "nightly" from the travis file. You can also remove ArrayViews from REQUIRE. I also have a branch EmpriricalRisks/rem_arrayviews that also replaces ArrayViews. (I've only made these updates locally as I plow through the package history for doing this kind of work. You might want to consider upgrading to newer, maintained packages) |
|
Okay thanks for the info. By the way, do you have a favorite replacement
package for high-performance L1-regularized linear regression? Thanks!
…On Mon, Mar 13, 2017 at 12:19 PM, Yaakov Borstein ***@***.***> wrote:
To get it through Travis you may want to remove the "nightly" from the
travis file.
You can also remove ArrayViews from REQUIRE.
I also have a branch EmpriricalRisks/rem_arrayviews that also replaces
ArrayViews.
(I've only made these updates locally as I plow through the package
history for doing this kind of work. You might want to consider upgrading
to newer, maintained packages)
—
You are receiving this because you modified the open/close state.
Reply to this email directly, view it on GitHub
<#18 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AE78hFQtY7KZbnhiZOrVHRc9qKpewg3_ks5rlXqggaJpZM4Lrj2E>
.
|
|
No but you can look at GLMNet, of which there are 2 versions, fortran based and pure Julia, Lasso.jl ... you could also look into the tool kits provided through JuliaML, JuMP ... where there is a ton of active development. |
Requires lindahua/EmpiricalRisks.jl#19