isogeny_class for EC_NumberField sometimes raises errors
54 views
Skip to first unread message
Barinder Banwait
Jul 12, 2022, 11:19:09 AM7/12/22
to sage-devel
Dear sage-devel,
I am running into an issue with computing isogeny classes of elliptic curves over number fields.
Here is what I entered:
K.<a> = QuadraticField(4569)
myJ = 46969655/32768
E = EllipticCurve(j=K(myJ))
C = E.isogeny_class()
This raises a KeyError in the first instance, which seems to be handled via a direct call to IsogenyClass_EC_NumberField, but this then raises a ValueError: Cannot convert infinity or NaN to Sage Integer. The full stack trace is in the PS.
This error has occurred on sage-9.4 on a system whose `uname -a` is `Linux LEGENDRE 5.4.0-91-generic #102-Ubuntu SMP Fri Nov 5 16:31:28 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux`, as well as on sage-9.7-beta5 on `Linux Barinder 5.4.0-121-generic #137-Ubuntu SMP Wed Jun 15 13:33:07 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux`.
Changing the field K to many other quadratic fields does not yield any error. However I did also notice this error with many j-invariants in QuadraticField(6537), so it somehow seems to be base-field dependent.
Best wishes,
Barinder
PS. Full traceback:
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/structure/category_object.pyx in sage.structure.category_object.CategoryObject.getattr_from_category (build/cythonized/sage/structure/category_object.c:7074)()
838 try:
--> 839 return self.__cached_methods[name]
840 except KeyError:
KeyError: '_isoclass'
During handling of the above exception, another exception occurred:
AttributeError Traceback (most recent call last)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogeny_class(self, reducible_primes, algorithm, minimal_models)
3007 try:
-> 3008 return self._isoclass
3009 except AttributeError:
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/structure/category_object.pyx in sage.structure.category_object.CategoryObject.__getattr__ (build/cythonized/sage/structure/category_object.c:6993)()
832 """
--> 833 return self.getattr_from_category(name)
834
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/structure/category_object.pyx in sage.structure.category_object.CategoryObject.getattr_from_category (build/cythonized/sage/structure/category_object.c:7159)()
847
--> 848 attr = getattr_from_other_class(self, cls, name)
849 self.__cached_methods[name] = attr
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/cpython/getattr.pyx in sage.cpython.getattr.getattr_from_other_class (build/cythonized/sage/cpython/getattr.c:2552)()
366 dummy_error_message.name = name
--> 367 raise AttributeError(dummy_error_message)
368 cdef PyObject* attr = instance_getattr(cls, name)
AttributeError: 'sage.sets.pythonclass.Set_PythonType_class' object has no attribute '_Hom_'
During handling of the above exception, another exception occurred:
ValueError Traceback (most recent call last)
<ipython-input-1-a9c20a51d066> in <module>
2 myJ = Integer(46969655)/Integer(32768)
3 E = EllipticCurve(j=K(myJ))
----> 4 C = E.isogeny_class()
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogeny_class(self, reducible_primes, algorithm, minimal_models)
3009 except AttributeError:
3010 from sage.schemes.elliptic_curves.isogeny_class import IsogenyClass_EC_NumberField
-> 3011 self._isoclass = IsogenyClass_EC_NumberField(self, reducible_primes=reducible_primes, algorithm=algorithm, minimal_models=minimal_models)
3012 return self._isoclass
3013
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_class.py in __init__(self, E, reducible_primes, algorithm, minimal_models)
705 self._algorithm = algorithm
706 self._minimal_models = minimal_models
--> 707 IsogenyClass_EC.__init__(self, E, label=None, empty=False)
708
709 def copy(self):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_class.py in __init__(self, E, label, empty)
68 self._label = label
69 if not empty:
---> 70 self._compute()
71
72 def __len__(self):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_class.py in _compute(self, verbose)
789 sys.stdout.write(" possible isogeny degrees: %s" % degs)
790 sys.stdout.flush()
--> 791 isogenies = E.isogenies_prime_degree(degs, minimal_models=self._minimal_models)
792 if verbose:
793 sys.stdout.write(" -actual isogeny degrees: %s" % Set(phi.degree() for phi in isogenies))
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogenies_prime_degree(self, l, algorithm, minimal_models)
3108 return self.isogenies_prime_degree(L, minimal_models=minimal_models)
3109
-> 3110 isogs = sum([self.isogenies_prime_degree(p, minimal_models=minimal_models) for p in l],
3111 [])
3112
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in <listcomp>(.0)
3108 return self.isogenies_prime_degree(L, minimal_models=minimal_models)
3109
-> 3110 isogs = sum([self.isogenies_prime_degree(p, minimal_models=minimal_models) for p in l],
3111 [])
3112
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogenies_prime_degree(self, l, algorithm, minimal_models)
3097 try:
3098 if l.is_prime(proof=False):
-> 3099 return isogenies_prime_degree(self, l, minimal_models=minimal_models)
3100 else:
3101 raise ValueError("%s is not prime." % l)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_small_degree.py in isogenies_prime_degree(E, l, minimal_models)
2340 return isogenies_2(E, minimal_models=minimal_models)
2341 if l==3:
-> 2342 return isogenies_3(E, minimal_models=minimal_models)
2343
2344 p = E.base_ring().characteristic()
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_small_degree.py in isogenies_3(E, minimal_models)
754 from sage.rings.number_field.number_field_base import is_NumberField
755 model = "minimal" if minimal_models and is_NumberField(E.base_field()) else None
--> 756 isogs = [E.isogeny(f, model=model) for f in ff]
757 return isogs
758
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_small_degree.py in <listcomp>(.0)
754 from sage.rings.number_field.number_field_base import is_NumberField
755 model = "minimal" if minimal_models and is_NumberField(E.base_field()) else None
--> 756 isogs = [E.isogeny(f, model=model) for f in ff]
757 return isogs
758
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_field.py in isogeny(self, kernel, codomain, degree, model, check)
876 """
877 try:
--> 878 return EllipticCurveIsogeny(self, kernel, codomain, degree, model, check=check)
879 except AttributeError as e:
880 raise RuntimeError("Unable to construct isogeny: %s" % e)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_curve_isogeny.py in __init__(self, E, kernel, codomain, degree, model, check)
1015 self.__compute_E2()
1016
-> 1017 self.__setup_post_isomorphism(codomain, model)
1018
1019 if (pre_isom is not None):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_curve_isogeny.py in __setup_post_isomorphism(self, codomain, model)
1800 raise ValueError("specifying minimal for model flag only valid with curves over number fields.")
1801
-> 1802 newE2 = oldE2.global_minimal_model(semi_global=True)
1803 post_isom = oldE2.isomorphism_to(newE2)
1804
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in global_minimal_model(self, proof, semi_global)
2042 from .kraus import semi_global_minimal_model
2043 E, P = semi_global_minimal_model(self)
-> 2044 return E._scale_by_units()._reduce_model()
2045
2046 raise ValueError("%s has no global minimal model! For a semi-global minimal model use semi_global=True" % self)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in _scale_by_units(self)
987 from sage.modules.all import vector
988 v = vector([(x4.abs().nth_root(4)+x6.abs().nth_root(6)).log()*d for x4,x6,d in zip(c4s,c6s,degs)])
--> 989 es = [e.round() for e in -Ainv*U*v]
990 u = prod([uj**ej for uj,ej in zip(fu,es)])
991 return self.scale_curve(u)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in <listcomp>(.0)
987 from sage.modules.all import vector
988 v = vector([(x4.abs().nth_root(4)+x6.abs().nth_root(6)).log()*d for x4,x6,d in zip(c4s,c6s,degs)])
--> 989 es = [e.round() for e in -Ainv*U*v]
990 u = prod([uj**ej for uj,ej in zip(fu,es)])
991 return self.scale_curve(u)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/rings/real_mpfr.pyx in sage.rings.real_mpfr.RealNumber.round (build/cythonized/sage/rings/real_mpfr.c:22391)()
3015 cdef RealNumber x = self._new()
3016 mpfr_round(x.value, self.value)
-> 3017 return x.integer_part()
3018
3019 def floor(self):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/rings/real_mpfr.pyx in sage.rings.real_mpfr.RealNumber.integer_part (build/cythonized/sage/rings/real_mpfr.c:17420)()
2256 """
2257 if not mpfr_number_p(self.value):
-> 2258 raise ValueError('Cannot convert infinity or NaN to Sage Integer')
2259
2260 cdef Integer z = Integer()
ValueError: Cannot convert infinity or NaN to Sage Integer
I am running into an issue with computing isogeny classes of elliptic curves over number fields.
Here is what I entered:
K.<a> = QuadraticField(4569)
myJ = 46969655/32768
E = EllipticCurve(j=K(myJ))
C = E.isogeny_class()
This raises a KeyError in the first instance, which seems to be handled via a direct call to IsogenyClass_EC_NumberField, but this then raises a ValueError: Cannot convert infinity or NaN to Sage Integer. The full stack trace is in the PS.
This error has occurred on sage-9.4 on a system whose `uname -a` is `Linux LEGENDRE 5.4.0-91-generic #102-Ubuntu SMP Fri Nov 5 16:31:28 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux`, as well as on sage-9.7-beta5 on `Linux Barinder 5.4.0-121-generic #137-Ubuntu SMP Wed Jun 15 13:33:07 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux`.
Changing the field K to many other quadratic fields does not yield any error. However I did also notice this error with many j-invariants in QuadraticField(6537), so it somehow seems to be base-field dependent.
Best wishes,
Barinder
PS. Full traceback:
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/structure/category_object.pyx in sage.structure.category_object.CategoryObject.getattr_from_category (build/cythonized/sage/structure/category_object.c:7074)()
838 try:
--> 839 return self.__cached_methods[name]
840 except KeyError:
KeyError: '_isoclass'
During handling of the above exception, another exception occurred:
AttributeError Traceback (most recent call last)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogeny_class(self, reducible_primes, algorithm, minimal_models)
3007 try:
-> 3008 return self._isoclass
3009 except AttributeError:
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/structure/category_object.pyx in sage.structure.category_object.CategoryObject.__getattr__ (build/cythonized/sage/structure/category_object.c:6993)()
832 """
--> 833 return self.getattr_from_category(name)
834
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/structure/category_object.pyx in sage.structure.category_object.CategoryObject.getattr_from_category (build/cythonized/sage/structure/category_object.c:7159)()
847
--> 848 attr = getattr_from_other_class(self, cls, name)
849 self.__cached_methods[name] = attr
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/cpython/getattr.pyx in sage.cpython.getattr.getattr_from_other_class (build/cythonized/sage/cpython/getattr.c:2552)()
366 dummy_error_message.name = name
--> 367 raise AttributeError(dummy_error_message)
368 cdef PyObject* attr = instance_getattr(cls, name)
AttributeError: 'sage.sets.pythonclass.Set_PythonType_class' object has no attribute '_Hom_'
During handling of the above exception, another exception occurred:
ValueError Traceback (most recent call last)
<ipython-input-1-a9c20a51d066> in <module>
2 myJ = Integer(46969655)/Integer(32768)
3 E = EllipticCurve(j=K(myJ))
----> 4 C = E.isogeny_class()
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogeny_class(self, reducible_primes, algorithm, minimal_models)
3009 except AttributeError:
3010 from sage.schemes.elliptic_curves.isogeny_class import IsogenyClass_EC_NumberField
-> 3011 self._isoclass = IsogenyClass_EC_NumberField(self, reducible_primes=reducible_primes, algorithm=algorithm, minimal_models=minimal_models)
3012 return self._isoclass
3013
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_class.py in __init__(self, E, reducible_primes, algorithm, minimal_models)
705 self._algorithm = algorithm
706 self._minimal_models = minimal_models
--> 707 IsogenyClass_EC.__init__(self, E, label=None, empty=False)
708
709 def copy(self):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_class.py in __init__(self, E, label, empty)
68 self._label = label
69 if not empty:
---> 70 self._compute()
71
72 def __len__(self):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_class.py in _compute(self, verbose)
789 sys.stdout.write(" possible isogeny degrees: %s" % degs)
790 sys.stdout.flush()
--> 791 isogenies = E.isogenies_prime_degree(degs, minimal_models=self._minimal_models)
792 if verbose:
793 sys.stdout.write(" -actual isogeny degrees: %s" % Set(phi.degree() for phi in isogenies))
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogenies_prime_degree(self, l, algorithm, minimal_models)
3108 return self.isogenies_prime_degree(L, minimal_models=minimal_models)
3109
-> 3110 isogs = sum([self.isogenies_prime_degree(p, minimal_models=minimal_models) for p in l],
3111 [])
3112
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in <listcomp>(.0)
3108 return self.isogenies_prime_degree(L, minimal_models=minimal_models)
3109
-> 3110 isogs = sum([self.isogenies_prime_degree(p, minimal_models=minimal_models) for p in l],
3111 [])
3112
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in isogenies_prime_degree(self, l, algorithm, minimal_models)
3097 try:
3098 if l.is_prime(proof=False):
-> 3099 return isogenies_prime_degree(self, l, minimal_models=minimal_models)
3100 else:
3101 raise ValueError("%s is not prime." % l)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_small_degree.py in isogenies_prime_degree(E, l, minimal_models)
2340 return isogenies_2(E, minimal_models=minimal_models)
2341 if l==3:
-> 2342 return isogenies_3(E, minimal_models=minimal_models)
2343
2344 p = E.base_ring().characteristic()
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_small_degree.py in isogenies_3(E, minimal_models)
754 from sage.rings.number_field.number_field_base import is_NumberField
755 model = "minimal" if minimal_models and is_NumberField(E.base_field()) else None
--> 756 isogs = [E.isogeny(f, model=model) for f in ff]
757 return isogs
758
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/isogeny_small_degree.py in <listcomp>(.0)
754 from sage.rings.number_field.number_field_base import is_NumberField
755 model = "minimal" if minimal_models and is_NumberField(E.base_field()) else None
--> 756 isogs = [E.isogeny(f, model=model) for f in ff]
757 return isogs
758
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_field.py in isogeny(self, kernel, codomain, degree, model, check)
876 """
877 try:
--> 878 return EllipticCurveIsogeny(self, kernel, codomain, degree, model, check=check)
879 except AttributeError as e:
880 raise RuntimeError("Unable to construct isogeny: %s" % e)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_curve_isogeny.py in __init__(self, E, kernel, codomain, degree, model, check)
1015 self.__compute_E2()
1016
-> 1017 self.__setup_post_isomorphism(codomain, model)
1018
1019 if (pre_isom is not None):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_curve_isogeny.py in __setup_post_isomorphism(self, codomain, model)
1800 raise ValueError("specifying minimal for model flag only valid with curves over number fields.")
1801
-> 1802 newE2 = oldE2.global_minimal_model(semi_global=True)
1803 post_isom = oldE2.isomorphism_to(newE2)
1804
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in global_minimal_model(self, proof, semi_global)
2042 from .kraus import semi_global_minimal_model
2043 E, P = semi_global_minimal_model(self)
-> 2044 return E._scale_by_units()._reduce_model()
2045
2046 raise ValueError("%s has no global minimal model! For a semi-global minimal model use semi_global=True" % self)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in _scale_by_units(self)
987 from sage.modules.all import vector
988 v = vector([(x4.abs().nth_root(4)+x6.abs().nth_root(6)).log()*d for x4,x6,d in zip(c4s,c6s,degs)])
--> 989 es = [e.round() for e in -Ainv*U*v]
990 u = prod([uj**ej for uj,ej in zip(fu,es)])
991 return self.scale_curve(u)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/schemes/elliptic_curves/ell_number_field.py in <listcomp>(.0)
987 from sage.modules.all import vector
988 v = vector([(x4.abs().nth_root(4)+x6.abs().nth_root(6)).log()*d for x4,x6,d in zip(c4s,c6s,degs)])
--> 989 es = [e.round() for e in -Ainv*U*v]
990 u = prod([uj**ej for uj,ej in zip(fu,es)])
991 return self.scale_curve(u)
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/rings/real_mpfr.pyx in sage.rings.real_mpfr.RealNumber.round (build/cythonized/sage/rings/real_mpfr.c:22391)()
3015 cdef RealNumber x = self._new()
3016 mpfr_round(x.value, self.value)
-> 3017 return x.integer_part()
3018
3019 def floor(self):
/scratch/sage/sage-9.4/local/lib/python3.8/site-packages/sage/rings/real_mpfr.pyx in sage.rings.real_mpfr.RealNumber.integer_part (build/cythonized/sage/rings/real_mpfr.c:17420)()
2256 """
2257 if not mpfr_number_p(self.value):
-> 2258 raise ValueError('Cannot convert infinity or NaN to Sage Integer')
2259
2260 cdef Integer z = Integer()
ValueError: Cannot convert infinity or NaN to Sage Integer
John Cremona
Jul 12, 2022, 12:11:19 PM7/12/22
to SAGE devel
Barinder, I'll look into this (tomorrow) -- it's my code. Please open
a ticket and CC me. It's happening in the code to reduce a model with
respect to scaling by units.
John
> --
> You received this message because you are subscribed to the Google Groups "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/6220eeaf-6fa3-450f-883b-4932f0cd2d74n%40googlegroups.com.
a ticket and CC me. It's happening in the code to reduce a model with
respect to scaling by units.
John
> You received this message because you are subscribed to the Google Groups "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/6220eeaf-6fa3-450f-883b-4932f0cd2d74n%40googlegroups.com.
Barinder Banwait
Jul 12, 2022, 2:41:43 PM7/12/22
to sage-devel
Thanks John -- ticket raised here: https://trac.sagemath.org/ticket/34174#ticket
John Cremona
Jul 13, 2022, 3:59:09 AM7/13/22
to SAGE devel
On Tue, 12 Jul 2022 at 19:41, Barinder Banwait <b.s.ba...@gmail.com> wrote:
>
> Thanks John -- ticket raised here: https://trac.sagemath.org/ticket/34174#ticket
As a workaround, you can do two things (which will not be necessary
>
> Thanks John -- ticket raised here: https://trac.sagemath.org/ticket/34174#ticket
after I have finished with that ticket):
(1) replace E by E.integral_model(), or (better)
E.global_minimal_model(semi_global=True) before computing the isogeny
class
(2) use minimal_models=False in calling the isogeny_class() method.
The problem with this specific quadratic field is that the fundamental
unit has one very small conjugate.
Further discussions can be on ticket 34174.
John
Reply all
Reply to author
Forward
0 new messages