-
Notifications
You must be signed in to change notification settings - Fork 65
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
Precision loss in p-adic linear algebra #1510
Comments
We can go a bit further and create an error
|
I guess in my case I want to call a division free algorithm ... but AA sees the padics as a field. |
Yes, unfortunately we don't have any useful linear algebra over inexact fields. Also, the whole model for p-adics, unramified extenions and arbitrary local fields is currently remodelled. In the meantime, you might make use of |
|
In this case, where the elementary divisors are all 1, the ring-approach together with the "Book" fixes will do the job neatly:
Da haette noch duth "s geteilt" werden muessen, da kommt das der Praezisionsverlusst rein |
@fieker interesting, I wasn't aware of |
I have a patched version of hnf/ snf that will produce transformation matrices without precisio loss, precisely for this purpose. It's still under discussion... |
Maybe the following example is not a bug ... but the solution computed is suboptimal
The following computations can be done modulo 29^2.
They should be possible with padics?
Similar, but with wierd printing
In some larger examples
solve
complains that no solution exists, although it clearly does.The text was updated successfully, but these errors were encountered: