Bug: "ValueError: array must not contain infs or NaNs"

[bmorris3]

Some values of input parameters give the following intimidating, cryptic and uninformative error:

(base) tycho:dot bmmorris$ python -c "from dot.tests.test_model import test_model_optimizer as f; f()"
optimizing logp for variables: [dot_R_spot, dot_lat, dot_lon, dot_comp_inc, dot_shear, dot_P_eq, dot_f0]
264it [00:06, 38.29it/s, logp=-6.690179e+04]array: [-0.15787993 -0.52647481 -0.84561983 -2.19530702 -0.25451239 -0.74975759
  0.59741809 -0.52742003 -0.4755868   0.75138306]
point: {'dot_f0_interval__': array(0.75138306), 'dot_P_eq_interval__': array(-0.4755868), 'dot_shear_interval__': array(-0.52742003), 'dot_comp_inc_interval__': array(0.59741809), 'dot_lon_interval__': array([[-0.25451239, -0.74975759]]), 'dot_lat_interval__': array([[-0.84561983, -2.19530702]]), 'dot_R_spot_interval__': array([[-0.15787993, -0.52647481]])}
Traceback (most recent call last):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 903, in __call__
    self.fn() if output_subset is None else\
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/op.py", line 892, in rval
    r = p(n, [x[0] for x in i], o)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/tensor/slinalg.py", line 250, in perform
    A, b, lower=True)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/linalg/basic.py", line 336, in solve_triangular
    b1 = _asarray_validated(b, check_finite=check_finite)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/_lib/_util.py", line 239, in _asarray_validated
    a = toarray(a)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/numpy/lib/function_base.py", line 486, in asarray_chkfinite
    "array must not contain infs or NaNs")
ValueError: array must not contain infs or NaNs

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/exoplanet/utils.py", line 197, in objective
    *get_args_for_theano_function(bij.rmap(vec), model=model)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 917, in __call__
    storage_map=getattr(self.fn, 'storage_map', None))
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/link.py", line 325, in raise_with_op
    reraise(exc_type, exc_value, exc_trace)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/six.py", line 702, in reraise
    raise value.with_traceback(tb)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 903, in __call__
    self.fn() if output_subset is None else\
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/op.py", line 892, in rval
    r = p(n, [x[0] for x in i], o)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/tensor/slinalg.py", line 250, in perform
    A, b, lower=True)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/linalg/basic.py", line 336, in solve_triangular
    b1 = _asarray_validated(b, check_finite=check_finite)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/_lib/_util.py", line 239, in _asarray_validated
    a = toarray(a)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/numpy/lib/function_base.py", line 486, in asarray_chkfinite
    "array must not contain infs or NaNs")
ValueError: array must not contain infs or NaNs
Apply node that caused the error: Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}(TensorConstant{[[1.287538..6413e-03]]}, Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)].0)
Toposort index: 65
Inputs types: [TensorType(float64, matrix), TensorType(float64, col)]
Inputs shapes: [(1418, 1418), (1418, 1)]
Inputs strides: [(8, 11344), (8, 8)]
Inputs values: ['not shown', 'not shown']
Outputs clients: [[InplaceDimShuffle{1,0}(Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}.0), Elemwise{mul,no_inplace}(TensorConstant{(1, 1) of 2.0}, TensorConstant{(1, 1) of 0.5}, Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}.0)]]

Backtrace when the node is created(use Theano flag traceback.limit=N to make it longer):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/distribution.py", line 84, in __new__
    return model.Var(name, dist, data, total_size, dims=dims)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/model.py", line 1118, in Var
    model=self,
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/model.py", line 1744, in __init__
    self.logp_sum_unscaledt = distribution.logp_sum(data)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/distribution.py", line 162, in logp_sum
    return tt.sum(self.logp(*args, **kwargs))
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 328, in logp
    quaddist, logdet, ok = self._quaddist(value)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 113, in _quaddist
    dist, logdet, ok = self._quaddist_cov(delta)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 138, in _quaddist_cov
    return self._quaddist_chol(delta)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 132, in _quaddist_chol
    delta_trans = self.solve_lower(chol_cov, delta.T).T

HINT: Use the Theano flag 'exception_verbosity=high' for a debugprint and storage map footprint of this apply node.
Traceback (most recent call last):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 903, in __call__
    self.fn() if output_subset is None else\
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/op.py", line 892, in rval
    r = p(n, [x[0] for x in i], o)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/tensor/slinalg.py", line 250, in perform
    A, b, lower=True)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/linalg/basic.py", line 336, in solve_triangular
    b1 = _asarray_validated(b, check_finite=check_finite)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/_lib/_util.py", line 239, in _asarray_validated
    a = toarray(a)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/numpy/lib/function_base.py", line 486, in asarray_chkfinite
    "array must not contain infs or NaNs")
ValueError: array must not contain infs or NaNs

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/Users/bmmorris/git/dot/dot/tests/test_model.py", line 24, in test_model_optimizer
    map_soln = m.optimize()
  File "/Users/bmmorris/git/dot/dot/model.py", line 309, in optimize
    map_soln = optimize(start=start, **kwargs)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/exoplanet/utils.py", line 218, in optimize
    info = minimize(objective, x0, **kwargs)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/_minimize.py", line 594, in minimize
    return _minimize_bfgs(fun, x0, args, jac, callback, **options)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 1016, in _minimize_bfgs
    old_fval, old_old_fval, amin=1e-100, amax=1e100)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 836, in _line_search_wolfe12
    **kwargs)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/linesearch.py", line 101, in line_search_wolfe1
    c1=c1, c2=c2, amax=amax, amin=amin, xtol=xtol)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/linesearch.py", line 175, in scalar_search_wolfe1
    phi1 = phi(stp)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/linesearch.py", line 87, in phi
    return f(xk + s*pk, *args)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 326, in function_wrapper
    return function(*(wrapper_args + args))
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 64, in __call__
    fg = self.fun(x, *args)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/exoplanet/utils.py", line 197, in objective
    *get_args_for_theano_function(bij.rmap(vec), model=model)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 917, in __call__
    storage_map=getattr(self.fn, 'storage_map', None))
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/link.py", line 325, in raise_with_op
    reraise(exc_type, exc_value, exc_trace)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/six.py", line 702, in reraise
    raise value.with_traceback(tb)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 903, in __call__
    self.fn() if output_subset is None else\
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/op.py", line 892, in rval
    r = p(n, [x[0] for x in i], o)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/tensor/slinalg.py", line 250, in perform
    A, b, lower=True)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/linalg/basic.py", line 336, in solve_triangular
    b1 = _asarray_validated(b, check_finite=check_finite)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/_lib/_util.py", line 239, in _asarray_validated
    a = toarray(a)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/numpy/lib/function_base.py", line 486, in asarray_chkfinite
    "array must not contain infs or NaNs")
ValueError: array must not contain infs or NaNs
Apply node that caused the error: Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}(TensorConstant{[[1.287538..6413e-03]]}, Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)].0)
Toposort index: 65
Inputs types: [TensorType(float64, matrix), TensorType(float64, col)]
Inputs shapes: [(1418, 1418), (1418, 1)]
Inputs strides: [(8, 11344), (8, 8)]
Inputs values: ['not shown', 'not shown']
Outputs clients: [[InplaceDimShuffle{1,0}(Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}.0), Elemwise{mul,no_inplace}(TensorConstant{(1, 1) of 2.0}, TensorConstant{(1, 1) of 0.5}, Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}.0)]]

Backtrace when the node is created(use Theano flag traceback.limit=N to make it longer):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/distribution.py", line 84, in __new__
    return model.Var(name, dist, data, total_size, dims=dims)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/model.py", line 1118, in Var
    model=self,
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/model.py", line 1744, in __init__
    self.logp_sum_unscaledt = distribution.logp_sum(data)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/distribution.py", line 162, in logp_sum
    return tt.sum(self.logp(*args, **kwargs))
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 328, in logp
    quaddist, logdet, ok = self._quaddist(value)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 113, in _quaddist
    dist, logdet, ok = self._quaddist_cov(delta)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 138, in _quaddist_cov
    return self._quaddist_chol(delta)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 132, in _quaddist_chol
    delta_trans = self.solve_lower(chol_cov, delta.T).T

HINT: Use the Theano flag 'exception_verbosity=high' for a debugprint and storage map footprint of this apply node.
265it [00:09, 29.26it/s, logp=-6.690179e+04]

[bmorris3]

Following the tip from the last "HINT": here's the debugprint from theano.

Debugprint of the apply node:
Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False} [id A] <TensorType(float64, col)> ''
 |TensorConstant{[[1.287538..6413e-03]]} [id B] <TensorType(float64, matrix)>
 |Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)] [id C] <TensorType(float64, col)> ''
   |TensorConstant{(1, 1) of 1.0} [id D] <TensorType(float64, (True, True))>
   |TensorConstant{[[0.975566..97422045]]} [id E] <TensorType(float64, col)>
   |InplaceDimShuffle{0,x} [id F] <TensorType(float64, col)> ''
   | |Sum{axis=[1], acc_dtype=float64} [id G] <TensorType(float64, vector)> ''
   |   |Elemwise{Composite{Switch(i0, (i1 * i2 * i3), i4)}} [id H] <TensorType(float64, matrix)> ''
   |     |Elemwise{Composite{GT((i0 - (sin((i1 - i2)) * i3 * i4)), i5)}} [id I] <TensorType(bool, matrix)> ''
   |     | |Elemwise{mul,no_inplace} [id J] <TensorType(float64, row)> ''
   |     | | |Elemwise{cos,no_inplace} [id K] <TensorType(float64, row)> ''
   |     | | | |Elemwise{Add}[(0, 1)] [id L] <TensorType(float64, row)> 'dot_lat'
   |     | | |InplaceDimShuffle{x,x} [id M] <TensorType(float64, (True, True))> ''
   |     | |   |Elemwise{sin,no_inplace} [id N] <TensorType(float64, scalar)> ''
   |     | |     |Elemwise{mul,no_inplace} [id O] <TensorType(float64, scalar)> 'dot_comp_inc'
   |     | |Elemwise{mul,no_inplace} [id P] <TensorType(float64, matrix)> ''
   |     | | |Elemwise{Composite{(i0 / (i1 / i2))}} [id Q] <TensorType(float64, row)> ''
   |     | | | |TensorConstant{(1, 1) of ..5307179586} [id R] <TensorType(float64, (True, True))>
   |     | | | |InplaceDimShuffle{x,x} [id S] <TensorType(float64, (True, True))> ''
   |     | | | | |Elemwise{add,no_inplace} [id T] <TensorType(float64, scalar)> 'dot_P_eq'
   |     | | | |Elemwise{Composite{(i0 - (i1 * i2))}} [id U] <TensorType(float64, row)> ''
   |     | | |   |TensorConstant{(1, 1) of 1.0} [id D] <TensorType(float64, (True, True))>
   |     | | |   |InplaceDimShuffle{x,x} [id V] <TensorType(float64, (True, True))> ''
   |     | | |   | |Elemwise{Composite{(i0 + (i1 * i2))}} [id W] <TensorType(float64, scalar)> 'dot_shear'
   |     | | |   |Elemwise{sqr,no_inplace} [id X] <TensorType(float64, row)> ''
   |     | | |     |Elemwise{sin,no_inplace} [id Y] <TensorType(float64, row)> ''
   |     | | |       |Elemwise{add,no_inplace} [id Z] <TensorType(float64, row)> ''
   |     | | |         |TensorConstant{(1, 1) of ..4015954636} [id BA] <TensorType(float64, (True, True))>
   |     | | |         |Elemwise{mul,no_inplace} [id BB] <TensorType(float64, row)> ''
   |     | | |           |TensorConstant{(1, 1) of ..8031909276} [id BC] <TensorType(float64, (True, True))>
   |     | | |           |sigmoid [id BD] <TensorType(float64, row)> ''
   |     | | |             |dot_lat_interval__ [id BE] <TensorType(float64, row)>
   |     | | |TensorConstant{[[-0.98859..01003145]]} [id BF] <TensorType(float64, col)>
   |     | |Elemwise{mul,no_inplace} [id BG] <TensorType(float64, row)> 'dot_lon'
   |     | |InplaceDimShuffle{x,x} [id BH] <TensorType(float64, (True, True))> ''
   |     | | |Elemwise{Cos}[(0, 0)] [id BI] <TensorType(float64, scalar)> ''
   |     | |   |Elemwise{mul,no_inplace} [id O] <TensorType(float64, scalar)> 'dot_comp_inc'
   |     | |Elemwise{Sin}[(0, 0)] [id BJ] <TensorType(float64, row)> ''
   |     | | |Elemwise{Add}[(0, 1)] [id L] <TensorType(float64, row)> 'dot_lat'
   |     | |TensorConstant{(1, 1) of 0} [id BK] <TensorType(int8, (True, True))>
   |     |TensorConstant{(1, 1) of 0.8} [id BL] <TensorType(float64, (True, True))>
   |     |Elemwise{sqr,no_inplace} [id BM] <TensorType(float64, row)> ''
   |     | |Elemwise{Composite{(i0 + (i1 * i2))}} [id BN] <TensorType(float64, row)> 'dot_R_spot'
   |     |Elemwise{Composite{sqrt((i0 - (sqr(i1) + sqr((-(i2 * i3))))))}} [id BO] <TensorType(float64, matrix)> ''
   |     | |TensorConstant{(1, 1) of 1.0} [id D] <TensorType(float64, (True, True))>
   |     | |Elemwise{Composite{((i0 * i1 * i2) + i3)}} [id BP] <TensorType(float64, matrix)> ''
   |     | | |Elemwise{cos} [id BQ] <TensorType(float64, matrix)> ''
   |     | | | |Elemwise{Composite{((i0 + i1) - i2)}}[(0, 1)] [id BR] <TensorType(float64, matrix)> ''
   |     | | |   |TensorConstant{(1, 1) of ..3267948966} [id BS] <TensorType(float64, (True, True))>
   |     | | |   |Elemwise{mul,no_inplace} [id P] <TensorType(float64, matrix)> ''
   |     | | |   |Elemwise{mul,no_inplace} [id BG] <TensorType(float64, row)> 'dot_lon'
   |     | | |InplaceDimShuffle{x,x} [id M] <TensorType(float64, (True, True))> ''
   |     | | |Elemwise{Sin}[(0, 0)] [id BJ] <TensorType(float64, row)> ''
   |     | | |Elemwise{mul,no_inplace} [id BT] <TensorType(float64, row)> ''
   |     | |   |InplaceDimShuffle{x,x} [id BH] <TensorType(float64, (True, True))> ''
   |     | |   |Elemwise{cos,no_inplace} [id K] <TensorType(float64, row)> ''
   |     | |Elemwise{Sin}[(0, 0)] [id BU] <TensorType(float64, matrix)> ''
   |     | | |Elemwise{Composite{((i0 + i1) - i2)}}[(0, 1)] [id BR] <TensorType(float64, matrix)> ''
   |     | |Elemwise{Sin}[(0, 0)] [id BJ] <TensorType(float64, row)> ''
   |     |TensorConstant{(1, 1) of 0} [id BK] <TensorType(int8, (True, True))>
   |InplaceDimShuffle{x,x} [id BV] <TensorType(float64, (True, True))> ''
     |Elemwise{mul,no_inplace} [id BW] <TensorType(float64, scalar)> ''
       |TensorConstant{3.0} [id BX] <TensorType(float64, scalar)>
       |sigmoid [id BY] <TensorType(float64, scalar)> ''
         |dot_f0_interval__ [id BZ] <TensorType(float64, scalar)>

Storage map footprint:
 - TensorConstant{[[1.287538..6413e-03]]}, Shape: (1418, 1418), ElemSize: 8 Byte(s), TotalSize: 16085792 Byte(s)
 - TensorConstant{[[1.287538..6413e-03]]}, Shape: (1418, 1418), ElemSize: 8 Byte(s), TotalSize: 16085792 Byte(s)
 - Elemwise{cos}.0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - Elemwise{Sin}[(0, 0)].0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - Elemwise{Composite{((i0 * i1 * i2) + i3)}}.0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - Elemwise{Composite{sqrt((i0 - (sqr(i1) + sqr((-(i2 * i3))))))}}.0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - TensorConstant{[[-0.98859..01003145]]}, Shape: (1418, 1), ElemSize: 8 Byte(s), TotalSize: 11344 Byte(s)
 - TensorConstant{[[0.975566..97422045]]}, Shape: (1418, 1), ElemSize: 8 Byte(s), TotalSize: 11344 Byte(s)
 - Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)].0, Shape: (1418, 1), ElemSize: 8 Byte(s), TotalSize: 11344 Byte(s)
 - Elemwise{Composite{GT((i0 - (sin((i1 - i2)) * i3 * i4)), i5)}}.0, Shape: (1418, 2), ElemSize: 1 Byte(s), TotalSize: 2836 Byte(s)
 - dot_lon_interval__, Input, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_lat_interval__, Input, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_R_spot_interval__, Input, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - sigmoid.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - sigmoid.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - sigmoid.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Alloc.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_lon, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_R_spot, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{add,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{sqr,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{sin,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{cos,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{Sin}[(0, 0)].0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{sqr,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{Composite{(i0 - (i1 * i2))}}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_f0_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_P_eq_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_shear_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_comp_inc_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{3.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..8031909276}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.5707963267948966}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..5307179586}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{1e-06}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.499999}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_shear, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.19999999999999996}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of 1e-06}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 0.499999}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4015954636}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.4515827052894548}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.4515827052894548}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{2.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - __logp_dot_comp_inc_interval__, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{sin,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.5}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.4}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_P_eq, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of 0.5}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..2519943295}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - Elemwise{Cos}[(0, 0)].0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.6}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..7283903604}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-1.6094379124341005}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.3817151341702488}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{2.7672931195787456}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-99.99999999999999}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.09999999999999998}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - __logp_dot_P_eq_interval__, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of 1.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..3267948966}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..0664093453}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 0.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..0664093453}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 2.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 0.8}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-1.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.0986123085021973}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.20016628700803937}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-1.8378770664093453}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - __logp_dot_f0_interval__, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of -0.8}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..1805619453}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..5597893728}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..9999999998}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-3.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{8411.25353783464}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..5307179586}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of -1.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4842090336}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4015954636}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..8031909276}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..0614359172}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..8501930168}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..2153316419}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..2476988255}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..7345693511}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-0.6931491805619453}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.3836465597893728}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-12.499999999999998}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of -12.499975}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of -1e-06}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of -0.499999}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4000000001}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-19.999999999999993}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.09999999999999998}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.19999999999999996}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-12.499975}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-1e-06}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.499999}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-6.283185307179586}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-inf}, Shape: (), ElemSize: 4 Byte(s), TotalSize: 4.0 Byte(s)
 - TensorConstant{(1, 1) of -inf}, Shape: (1, 1), ElemSize: 4 Byte(s), TotalSize: 4 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (1, 2), ElemSize: 1 Byte(s), TotalSize: 2 Byte(s)
 - TensorConstant{1}, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - TensorConstant{0}, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - Elemwise{Composite{Cast{int8}(GE(i0, i1))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - TensorConstant{(1, 1) of 0}, Shape: (1, 1), ElemSize: 1 Byte(s), TotalSize: 1 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 TotalSize: 32300173.0 Byte(s) 0.030 GB
 TotalSize inputs: 32194899.0 Byte(s) 0.030 GB


Traceback (most recent call last):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 903, in __call__
    self.fn() if output_subset is None else\
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/op.py", line 892, in rval
    r = p(n, [x[0] for x in i], o)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/tensor/slinalg.py", line 250, in perform
    A, b, lower=True)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/linalg/basic.py", line 336, in solve_triangular
    b1 = _asarray_validated(b, check_finite=check_finite)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/_lib/_util.py", line 239, in _asarray_validated
    a = toarray(a)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/numpy/lib/function_base.py", line 486, in asarray_chkfinite
    "array must not contain infs or NaNs")
ValueError: array must not contain infs or NaNs

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/Users/bmmorris/git/dot/dot/tests/test_model.py", line 26, in test_model_optimizer
    map_soln = m.optimize()
  File "/Users/bmmorris/git/dot/dot/model.py", line 309, in optimize
    map_soln = optimize(start=start, **kwargs)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/exoplanet/utils.py", line 218, in optimize
    info = minimize(objective, x0, **kwargs)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/_minimize.py", line 594, in minimize
    return _minimize_bfgs(fun, x0, args, jac, callback, **options)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 1016, in _minimize_bfgs
    old_fval, old_old_fval, amin=1e-100, amax=1e100)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 836, in _line_search_wolfe12
    **kwargs)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/linesearch.py", line 101, in line_search_wolfe1
    c1=c1, c2=c2, amax=amax, amin=amin, xtol=xtol)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/linesearch.py", line 175, in scalar_search_wolfe1
    phi1 = phi(stp)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/linesearch.py", line 87, in phi
    return f(xk + s*pk, *args)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 326, in function_wrapper
    return function(*(wrapper_args + args))
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/optimize/optimize.py", line 64, in __call__
    fg = self.fun(x, *args)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/exoplanet/utils.py", line 197, in objective
    *get_args_for_theano_function(bij.rmap(vec), model=model)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 917, in __call__
    storage_map=getattr(self.fn, 'storage_map', None))
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/link.py", line 325, in raise_with_op
    reraise(exc_type, exc_value, exc_trace)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/six.py", line 702, in reraise
    raise value.with_traceback(tb)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/compile/function_module.py", line 903, in __call__
    self.fn() if output_subset is None else\
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/gof/op.py", line 892, in rval
    r = p(n, [x[0] for x in i], o)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/theano/tensor/slinalg.py", line 250, in perform
    A, b, lower=True)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/linalg/basic.py", line 336, in solve_triangular
    b1 = _asarray_validated(b, check_finite=check_finite)
  File "/Users/bmmorris/.local/lib/python3.7/site-packages/scipy-1.3.0-py3.7-macosx-10.9-x86_64.egg/scipy/_lib/_util.py", line 239, in _asarray_validated
    a = toarray(a)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/numpy/lib/function_base.py", line 486, in asarray_chkfinite
    "array must not contain infs or NaNs")
ValueError: array must not contain infs or NaNs
Apply node that caused the error: Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}(TensorConstant{[[1.287538..6413e-03]]}, Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)].0)
Toposort index: 65
Inputs types: [TensorType(float64, matrix), TensorType(float64, col)]
Inputs shapes: [(1418, 1418), (1418, 1)]
Inputs strides: [(8, 11344), (8, 8)]
Inputs values: ['not shown', 'not shown']
Inputs type_num: [12, 12]
Outputs clients: [[InplaceDimShuffle{1,0}(Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}.0), Elemwise{mul,no_inplace}(TensorConstant{(1, 1) of 2.0}, TensorConstant{(1, 1) of 0.5}, Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False}.0)]]

Backtrace when the node is created(use Theano flag traceback.limit=N to make it longer):
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/distribution.py", line 84, in __new__
    return model.Var(name, dist, data, total_size, dims=dims)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/model.py", line 1118, in Var
    model=self,
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/model.py", line 1744, in __init__
    self.logp_sum_unscaledt = distribution.logp_sum(data)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/distribution.py", line 162, in logp_sum
    return tt.sum(self.logp(*args, **kwargs))
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 328, in logp
    quaddist, logdet, ok = self._quaddist(value)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 113, in _quaddist
    dist, logdet, ok = self._quaddist_cov(delta)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 138, in _quaddist_cov
    return self._quaddist_chol(delta)
  File "/Users/bmmorris/miniconda3/lib/python3.7/site-packages/pymc3/distributions/multivariate.py", line 132, in _quaddist_chol
    delta_trans = self.solve_lower(chol_cov, delta.T).T

Debugprint of the apply node:
Solve{A_structure='lower_triangular', lower=False, overwrite_A=False, overwrite_b=False} [id A] <TensorType(float64, col)> ''
 |TensorConstant{[[1.287538..6413e-03]]} [id B] <TensorType(float64, matrix)>
 |Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)] [id C] <TensorType(float64, col)> ''
   |TensorConstant{(1, 1) of 1.0} [id D] <TensorType(float64, (True, True))>
   |TensorConstant{[[0.975566..97422045]]} [id E] <TensorType(float64, col)>
   |InplaceDimShuffle{0,x} [id F] <TensorType(float64, col)> ''
   | |Sum{axis=[1], acc_dtype=float64} [id G] <TensorType(float64, vector)> ''
   |   |Elemwise{Composite{Switch(i0, (i1 * i2 * i3), i4)}} [id H] <TensorType(float64, matrix)> ''
   |     |Elemwise{Composite{GT((i0 - (sin((i1 - i2)) * i3 * i4)), i5)}} [id I] <TensorType(bool, matrix)> ''
   |     | |Elemwise{mul,no_inplace} [id J] <TensorType(float64, row)> ''
   |     | | |Elemwise{cos,no_inplace} [id K] <TensorType(float64, row)> ''
   |     | | | |Elemwise{Add}[(0, 1)] [id L] <TensorType(float64, row)> 'dot_lat'
   |     | | |InplaceDimShuffle{x,x} [id M] <TensorType(float64, (True, True))> ''
   |     | |   |Elemwise{sin,no_inplace} [id N] <TensorType(float64, scalar)> ''
   |     | |     |Elemwise{mul,no_inplace} [id O] <TensorType(float64, scalar)> 'dot_comp_inc'
   |     | |Elemwise{mul,no_inplace} [id P] <TensorType(float64, matrix)> ''
   |     | | |Elemwise{Composite{(i0 / (i1 / i2))}} [id Q] <TensorType(float64, row)> ''
   |     | | | |TensorConstant{(1, 1) of ..5307179586} [id R] <TensorType(float64, (True, True))>
   |     | | | |InplaceDimShuffle{x,x} [id S] <TensorType(float64, (True, True))> ''
   |     | | | | |Elemwise{add,no_inplace} [id T] <TensorType(float64, scalar)> 'dot_P_eq'
   |     | | | |Elemwise{Composite{(i0 - (i1 * i2))}} [id U] <TensorType(float64, row)> ''
   |     | | |   |TensorConstant{(1, 1) of 1.0} [id D] <TensorType(float64, (True, True))>
   |     | | |   |InplaceDimShuffle{x,x} [id V] <TensorType(float64, (True, True))> ''
   |     | | |   | |Elemwise{Composite{(i0 + (i1 * i2))}} [id W] <TensorType(float64, scalar)> 'dot_shear'
   |     | | |   |Elemwise{sqr,no_inplace} [id X] <TensorType(float64, row)> ''
   |     | | |     |Elemwise{sin,no_inplace} [id Y] <TensorType(float64, row)> ''
   |     | | |       |Elemwise{add,no_inplace} [id Z] <TensorType(float64, row)> ''
   |     | | |         |TensorConstant{(1, 1) of ..4015954636} [id BA] <TensorType(float64, (True, True))>
   |     | | |         |Elemwise{mul,no_inplace} [id BB] <TensorType(float64, row)> ''
   |     | | |           |TensorConstant{(1, 1) of ..8031909276} [id BC] <TensorType(float64, (True, True))>
   |     | | |           |sigmoid [id BD] <TensorType(float64, row)> ''
   |     | | |             |dot_lat_interval__ [id BE] <TensorType(float64, row)>
   |     | | |TensorConstant{[[-0.98859..01003145]]} [id BF] <TensorType(float64, col)>
   |     | |Elemwise{mul,no_inplace} [id BG] <TensorType(float64, row)> 'dot_lon'
   |     | |InplaceDimShuffle{x,x} [id BH] <TensorType(float64, (True, True))> ''
   |     | | |Elemwise{Cos}[(0, 0)] [id BI] <TensorType(float64, scalar)> ''
   |     | |   |Elemwise{mul,no_inplace} [id O] <TensorType(float64, scalar)> 'dot_comp_inc'
   |     | |Elemwise{Sin}[(0, 0)] [id BJ] <TensorType(float64, row)> ''
   |     | | |Elemwise{Add}[(0, 1)] [id L] <TensorType(float64, row)> 'dot_lat'
   |     | |TensorConstant{(1, 1) of 0} [id BK] <TensorType(int8, (True, True))>
   |     |TensorConstant{(1, 1) of 0.8} [id BL] <TensorType(float64, (True, True))>
   |     |Elemwise{sqr,no_inplace} [id BM] <TensorType(float64, row)> ''
   |     | |Elemwise{Composite{(i0 + (i1 * i2))}} [id BN] <TensorType(float64, row)> 'dot_R_spot'
   |     |Elemwise{Composite{sqrt((i0 - (sqr(i1) + sqr((-(i2 * i3))))))}} [id BO] <TensorType(float64, matrix)> ''
   |     | |TensorConstant{(1, 1) of 1.0} [id D] <TensorType(float64, (True, True))>
   |     | |Elemwise{Composite{((i0 * i1 * i2) + i3)}} [id BP] <TensorType(float64, matrix)> ''
   |     | | |Elemwise{cos} [id BQ] <TensorType(float64, matrix)> ''
   |     | | | |Elemwise{Composite{((i0 + i1) - i2)}}[(0, 1)] [id BR] <TensorType(float64, matrix)> ''
   |     | | |   |TensorConstant{(1, 1) of ..3267948966} [id BS] <TensorType(float64, (True, True))>
   |     | | |   |Elemwise{mul,no_inplace} [id P] <TensorType(float64, matrix)> ''
   |     | | |   |Elemwise{mul,no_inplace} [id BG] <TensorType(float64, row)> 'dot_lon'
   |     | | |InplaceDimShuffle{x,x} [id M] <TensorType(float64, (True, True))> ''
   |     | | |Elemwise{Sin}[(0, 0)] [id BJ] <TensorType(float64, row)> ''
   |     | | |Elemwise{mul,no_inplace} [id BT] <TensorType(float64, row)> ''
   |     | |   |InplaceDimShuffle{x,x} [id BH] <TensorType(float64, (True, True))> ''
   |     | |   |Elemwise{cos,no_inplace} [id K] <TensorType(float64, row)> ''
   |     | |Elemwise{Sin}[(0, 0)] [id BU] <TensorType(float64, matrix)> ''
   |     | | |Elemwise{Composite{((i0 + i1) - i2)}}[(0, 1)] [id BR] <TensorType(float64, matrix)> ''
   |     | |Elemwise{Sin}[(0, 0)] [id BJ] <TensorType(float64, row)> ''
   |     |TensorConstant{(1, 1) of 0} [id BK] <TensorType(int8, (True, True))>
   |InplaceDimShuffle{x,x} [id BV] <TensorType(float64, (True, True))> ''
     |Elemwise{mul,no_inplace} [id BW] <TensorType(float64, scalar)> ''
       |TensorConstant{3.0} [id BX] <TensorType(float64, scalar)>
       |sigmoid [id BY] <TensorType(float64, scalar)> ''
         |dot_f0_interval__ [id BZ] <TensorType(float64, scalar)>

Storage map footprint:
 - TensorConstant{[[1.287538..6413e-03]]}, Shape: (1418, 1418), ElemSize: 8 Byte(s), TotalSize: 16085792 Byte(s)
 - TensorConstant{[[1.287538..6413e-03]]}, Shape: (1418, 1418), ElemSize: 8 Byte(s), TotalSize: 16085792 Byte(s)
 - Elemwise{cos}.0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - Elemwise{Sin}[(0, 0)].0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - Elemwise{Composite{((i0 * i1 * i2) + i3)}}.0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - Elemwise{Composite{sqrt((i0 - (sqr(i1) + sqr((-(i2 * i3))))))}}.0, Shape: (1418, 2), ElemSize: 8 Byte(s), TotalSize: 22688 Byte(s)
 - TensorConstant{[[-0.98859..01003145]]}, Shape: (1418, 1), ElemSize: 8 Byte(s), TotalSize: 11344 Byte(s)
 - TensorConstant{[[0.975566..97422045]]}, Shape: (1418, 1), ElemSize: 8 Byte(s), TotalSize: 11344 Byte(s)
 - Elemwise{Composite{((i0 + i1 + i2) - i3)}}[(0, 2)].0, Shape: (1418, 1), ElemSize: 8 Byte(s), TotalSize: 11344 Byte(s)
 - Elemwise{Composite{GT((i0 - (sin((i1 - i2)) * i3 * i4)), i5)}}.0, Shape: (1418, 2), ElemSize: 1 Byte(s), TotalSize: 2836 Byte(s)
 - dot_lon_interval__, Input, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_lat_interval__, Input, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_R_spot_interval__, Input, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - sigmoid.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - sigmoid.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - sigmoid.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Alloc.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_lon, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_R_spot, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{add,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{sqr,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{sin,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{cos,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{Sin}[(0, 0)].0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{sqr,no_inplace}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - Elemwise{Composite{(i0 - (i1 * i2))}}.0, Shape: (1, 2), ElemSize: 8 Byte(s), TotalSize: 16 Byte(s)
 - dot_f0_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_P_eq_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_shear_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_comp_inc_interval__, Input, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - sigmoid.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{neg,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{3.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..8031909276}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.5707963267948966}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..5307179586}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{1e-06}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.499999}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_shear, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.19999999999999996}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of 1e-06}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 0.499999}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4015954636}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.4515827052894548}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.4515827052894548}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{2.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - __logp_dot_comp_inc_interval__, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - Elemwise{sin,no_inplace}.0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.5}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.4}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - dot_P_eq, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of 0.5}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..2519943295}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - Elemwise{Cos}[(0, 0)].0, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{0.6}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..7283903604}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - InplaceDimShuffle{x,x}.0, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-1.6094379124341005}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.3817151341702488}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{2.7672931195787456}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-99.99999999999999}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.09999999999999998}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - __logp_dot_P_eq_interval__, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of 1.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..3267948966}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..0664093453}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 0.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..0664093453}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 2.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of 0.8}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-1.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.0986123085021973}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.20016628700803937}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-1.8378770664093453}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - __logp_dot_f0_interval__, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of -0.8}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..1805619453}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..5597893728}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..9999999998}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-3.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.0}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{8411.25353783464}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of ..5307179586}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of -1.0}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4842090336}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4015954636}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..8031909276}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..0614359172}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..8501930168}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..2153316419}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..2476988255}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..7345693511}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-0.6931491805619453}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{1.3836465597893728}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-12.499999999999998}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{(1, 1) of -12.499975}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of -1e-06}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of -0.499999}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{(1, 1) of ..4000000001}, Shape: (1, 1), ElemSize: 8 Byte(s), TotalSize: 8 Byte(s)
 - TensorConstant{-19.999999999999993}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{0.09999999999999998}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.19999999999999996}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-12.499975}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-1e-06}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-0.499999}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-6.283185307179586}, Shape: (), ElemSize: 8 Byte(s), TotalSize: 8.0 Byte(s)
 - TensorConstant{-inf}, Shape: (), ElemSize: 4 Byte(s), TotalSize: 4.0 Byte(s)
 - TensorConstant{(1, 1) of -inf}, Shape: (1, 1), ElemSize: 4 Byte(s), TotalSize: 4 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (1, 2), ElemSize: 1 Byte(s), TotalSize: 2 Byte(s)
 - TensorConstant{1}, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - TensorConstant{0}, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - Elemwise{Composite{Cast{int8}(GE(i0, i1))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - TensorConstant{(1, 1) of 0}, Shape: (1, 1), ElemSize: 1 Byte(s), TotalSize: 1 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 - Elemwise{Composite{Cast{int8}((GE(i0, i1) * LE(i0, i2)))}}.0, Shape: (), ElemSize: 1 Byte(s), TotalSize: 1.0 Byte(s)
 TotalSize: 32300173.0 Byte(s) 0.030 GB
 TotalSize inputs: 32194899.0 Byte(s) 0.030 GB

[bmorris3]

After a really deep dive, the problem seems to be that we define the rotation period of individual spots with a division that could symbolically produce infs. To avoid this, we can wrap the spot_period expression in a pm.math.where call which turns off the calculation when the shear -> 1, bypassing the problem.

[bmorris3]

This issue is reoccurring when n_spots>6 for Kepler-17. Fix incoming.

[bmorris3]

This is a cruel, cruel world. Seems this error is also being caused by non-matching datatypes.

[bmorris3]

Another possible lead: if there's more than one numpy installed that can cause issues.