Make BooleanPolynomial.variables() way faster #35510
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The implementation of `BooleanMonomialVariableIterator` is a bit slow.
It is based in polybori's `PBMonomVarIter` which implements an iterator
over the variables of a boolean monomial which is slow.
However there is `PBMonomIter` which implements an iterator over the
indices of the variables which seems to be faster; the variables can
then be retrieved from the ring itself.
This is what we do in this commit. Performance is way better.
I don't know why this difference, maybe because the variables are strings
and this suffers from allocation, while indices are just 32 bit ints.
Before this commit:
```
sage: R = BooleanPolynomialRing(512)
sage: f = sum(R.gen(randrange(512)) for _ in range(20))
sage: timeit('vs = f.variables()')
125 loops, best of 3: 6.1 ms per loop
```
After this commit:
```
sage: R = BooleanPolynomialRing(512)
sage: f = sum(R.gen(randrange(512)) for _ in range(20))
sage: timeit('vs = f.variables()')
625 loops, best of 3: 13.7 μs per loop
```
Another example taken from a doctest for the method
`PolynomialSequence_generic.connected_components()` in
`src/sage/rings/polynomial/multi_polynomial_sequence.py`:
Before this commit:
```
sage: sr = mq.SR(2,4,4,8,gf2=True,polybori=True)
sage: F, s = sr.polynomial_system()
sage: Fz = Sequence(F.part(2))
sage: %time vs = Fz.variables()
CPU times: user 603 ms, sys: 2.42 s, total: 3.03 s
Wall time: 3.03 s
sage: %time len(Fz.connected_components())
CPU times: user 2.59 s, sys: 10.9 s, total: 13.5 s
Wall time: 13.5 s
4
```
After this commit:
```
sage: sr = mq.SR(2,4,4,8,gf2=True,polybori=True)
sage: F, s = sr.polynomial_system()
sage: Fz = Sequence(F.part(2))
sage: %time vs = Fz.variables()
CPU times: user 9.77 ms, sys: 994 µs, total: 10.8 ms
Wall time: 13.7 ms
sage: %time len(Fz.connected_components())
CPU times: user 70.8 ms, sys: 8.27 ms, total: 79.1 ms
Wall time: 82.5 ms
4
```
tornaria
added a commit
to tornaria/sage
that referenced
this pull request
Apr 15, 2023
Simplify the code; also, the previous code has to iterate over variables of the sequence twice (this is really bad before sagemath#35510)
3 tasks
|
Documentation preview for this PR is ready! 🎉 |
videlec
approved these changes
Apr 15, 2023
tornaria
added a commit
to tornaria/sage
that referenced
this pull request
Apr 15, 2023
Simplify the code; also, the previous code has to iterate over variables of the sequence twice (this is really bad before sagemath#35510) Moreover, now we add a clique between the variables of each polynomial, so it agrees with the description (the code before used to add just a spanning tree of the clique -- a star). This makes this method a little bit slower for the purposes of `connected_components()` (for which adding a star is equivalent). However, sagemath#35518 will rewrite `connected_components()` without using `connection_graph()` so this is not really a problem.
vbraun
pushed a commit
that referenced
this pull request
Apr 23, 2023
Simplify the code; also, the previous code has to iterate over variables of the sequence twice (this is really bad before #35510). This affects mainly the method `connected_components()`. EDIT: Moreover, now we add a clique between the variables of each polynomial, so it agrees with the description (the code before used to add just a spanning tree of the clique -- a star). This makes this method a little bit slower for the purposes of `connected_components()` (for which adding a star is equivalent). However, #35518 will rewrite `connected_components()` without using `connection_graph()` so this is not really a problem. ### 📝 Checklist - [x] The title is concise, informative, and self-explanatory. - [x] The description explains in detail what this PR is about. - [x] New tests added to check that a clique is added for the variables in each polynomial. ### Dependencies - #35511 URL: #35512 Reported by: Gonzalo Tornaría Reviewer(s): David Coudert, Gonzalo Tornaría, Vincent Delecroix
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
📚 Description
The implementation of
BooleanMonomialVariableIteratoris a bit slow. It is based in polybori'sPBMonomVarIterwhich implements an iterator over the variables of a boolean monomial which is slow.However there is
PBMonomIterwhich implements an iterator over the indices of the variables which seems to be faster; the variables can then be retrieved from the ring itself.This is what we do in this commit. Performance is way better.
I don't know why this difference, maybe because the variables are strings and this suffers from allocation, while indices are just 32 bit ints.
Before this commit:
After this commit:
Another example taken from a doctest for the method
PolynomialSequence_generic.connected_components()insrc/sage/rings/polynomial/multi_polynomial_sequence.py:Before this commit:
After this commit:
📝 Checklist