SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1604.05723 · 2016

ALMA Imaging and Gravitational Lens Models of South Pole Telescope-Selected Dusty, Star-Forming Galaxies at High Redshifts

arXiv · PDF · Open in the Atlas

Code that ran

7 of 14 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 0 failures of our instrument; 7 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1604.05723.json · badge: /api/ran/1604.05723.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
jspilker/visilens canonical (classifier label)
classifier basis: census
7 ran · 7 no recorded run
FunctionStatusWhere it lives
GenerateLensingGrid Ran
ran on a synthesized input
jspilker/visilens/visilens/lensing.py
code served (permissive licence) · get_code("aa7d776e7198b253")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

class MockVisData:
    """Mock visibility data object with uvdist attribute"""
    def __init__(self, uvdist_max=1000.0):
        self.uvdist = np.linspace(0, uvdist_max, 100)

# Create mock data object
mock_vis = MockVisData(uvdist_max=1500.0)

ARGS = {
    "data": mock_vis,
    "xmax": 30.0,
    "emissionbox": [-5, 5, -5, 5],
    "fieldres": None,
    "emitres": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
LensRayTrace Ran
ran on a synthesized input
jspilker/visilens/visilens/lensing.py
code served (permissive licence) · get_code("638d530755e28947")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

class SIELens:
    """Stand-in for SIE (Singular Isothermal Ellipsoid) lens"""
    def __init__(self):
        self.deflected_x = np.zeros((2, 4))
        self.deflected_y = np.zeros((2, 4))
    
    def deflect(self, xim, yim, Dd, Ds, Dds):
        """Mock deflection calculation"""
        self.deflected_x = np.random.randn(2, 4) * 0.1
        self.deflected_y = np.random.randn(2, 4) * 0.1

class ExternalShear:
    """Stand-in for external shear"""
    def __init__(self):
        self.deflected_x = np.zeros((2, 4))
        self.deflected_y = np.zeros((2, 4))
    
    def deflect(self, xim, yim, ref_lens):
        """Mock shear deflection"""
        self.deflected_x = np.random.randn(2, 4) * 0.05
        self.deflected_y = np.random.randn(2, 4) * 0.05

ARGS = {
    'xim': np.random.randn(2, 4),
    'yim': np.random.randn(2, 4),
    'lens': [SIELens(), ExternalShear()],
    'Dd': 1000.0,
    'Ds': 1500.0,
    'Dds': 500.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
cart2pol Ran
ran on a synthesized input
jspilker/visilens/visilens/utils.py
code served (permissive licence) · get_code("c7a20bf2893441bc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small test arrays for x and y coordinates
x = np.array([1.0, 2.0, 3.0, 4.0])
y = np.array([1.0, 2.0, 3.0, 4.0])

ARGS = {
    'x': x,
    'y': y
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
concatvis Ran
ran on a synthesized input
jspilker/visilens/visilens/class_utils.py
code served (permissive licence) · get_code("e0082627f29e9fd3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Visdata. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small visibility data objects
def create_visdata(n_vis=10):
    u = np.random.randn(n_vis).astype(np.float32)
    v = np.random.randn(n_vis).astype(np.float32)
    real = np.random.randn(n_vis).astype(np.float32)
    imag = np.random.randn(n_vis).astype(np.float32)
    sigma = np.ones(n_vis, dtype=np.float32) * 0.1
    ant1 = np.arange(n_vis, dtype=np.int32) % 4
    ant2 = np.arange(n_vis, dtype=np.int32) % 3
    
    class Visdata(object):
        def __init__(self, u, v, real, imag, sigma, ant1=None, ant2=None, PBfwhm=None, filename=None):
            self.u = u
            self.v = v
            self.real = real
            self.imag = imag
            self.sigma = sigma
            self.ant1 = ant1
            self.ant2 = ant2
            self.PBfwhm = PBfwhm
            self.filename = filename

        @property
        def uvdist(self):
            return np.sqrt(self.u ** 2.0 + self.v ** 2.0)

        @property
        def real(self):
            return self._real

        @real.setter
        def real(self, val):
            self._real = val
            try:
                self._amp = np.sqrt(self._real ** 2.0 + self.imag ** 2.0)
                self._phase = np.arctan2(self.imag, self._real)
            except (AttributeError, ValueError):
                self._amp = None
                self._phase = None

        @property
        def imag(self):
            return self._imag

        @imag.setter
        def imag(self, val):
            self._imag = val
            try:
                self._amp = np.sqrt(self.real ** 2.0 + self._imag ** 2.0)
                self._phase = np.arctan2(self._imag, self.real)
            except (AttributeError, ValueError):
                self._amp = None
                self._phase = None

        @property
        def amp(self):
            return self._amp

        @amp.setter
        def amp(self, val):
            self._amp = val
            self._real = val * np.cos(self.phase)
            self._imag = val * np.sin(self.phase)

        @property
        def phase(self):
            return self._phase

        @phase.setter
        def phase(self, val):
            self._phase = val
            self._real = self.amp * np.cos(val)
            self._imag = self.amp * np.sin(val)

        def __add__(self, other):
            return Visdata(self.u, self.v, self.real + other.real, self.imag + other.imag, (self.sigma ** (-2.0) + other.sigma ** (-2.0)) ** (-0.5))

        def __sub__(self, other):
            return Visdata(self.u, self.v, self.real - other.real, self.imag - other.imag, (self.sigma ** (-2.0) + other.sigma ** (-2.0)) ** (-0.5))

        def conjugate(self):
            u = np.concatenate((self.u, -self.u))
            v = np.concatenate((self.v, -self.v))
            real = np.concatenate((self.real, self.real))
            imag = np.concatenate((self.imag, -self.imag))
            sigma = np.concatenate((self.sigma, self.sigma))
            ant1 = np.concatenate((self.ant1, self.ant2))
            ant2 = np.concatenate((self.ant2, self.ant1))
            self.u = u
            self.v = v
            self.real = real
            self.imag = imag
            self.sigma = sigma
            self.ant1 = ant1
            self.ant2 = ant2
    
    return Visdata(u, v, real, imag, sigma, ant1, ant2, PBfwhm=0.5, filename="test.bin")

# Create multiple visibility objects
vis1 = create_visdata(n_vis=8)
vis2 = create_visdata(n_vis=6)
vis3 = create_visdata(n_vis=7)

visdatas = [vis1, vis2, vis3]

ARGS = {
    "visdatas": visdatas
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Visdata — a Visdata output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Visdata output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pass_priors Ran
ran on a synthesized input
jspilker/visilens/visilens/calc_likelihood.py
code served (permissive licence) · get_code("9c8471f16830c9ae")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import copy

class ParameterDict:
    """Stand-in for parameter objects with prior and fixed attributes"""
    def __init__(self, value, prior_min, prior_max, fixed=False):
        self.value = value
        self.prior = (prior_min, prior_max)
        self.fixed = fixed
    
    def __getitem__(self, key):
        if key == 'value':
            return self.value
        elif key == 'prior':
            return self.prior
        elif key == 'fixed':
            return self.fixed
        raise KeyError(key)
    
    def __setitem__(self, key, val):
        if key == 'value':
            self.value = val
        elif key == 'prior':
            self.prior = val
        elif key == 'fixed':
            self.fixed = val

class SIELens:
    """Stand-in SIE lens object"""
    def __init__(self):
        self.__class__.__name__ = 'SIELens'
        self._altered = False
        self.x = ParameterDict(0.5, 0.0, 1.0, fixed=False)
        self.y = ParameterDict(0.5, 0.0, 1.0, fixed=False)
        self.M = ParameterDict(1.0, 0.5, 2.0, fixed=False)
        self.e = ParameterDict(0.3, 0.0, 0.5, fixed=True)
        self.PA = ParameterDict(45.0, 0.0, 180.0, fixed=True)

class ExternalShear:
    """Stand-in external shear object"""
    def __init__(self):
        self.__class__.__name__ = 'ExternalShear'
        self._altered = False
        self.shear = ParameterDict(0.05, 0.0, 0.1, fixed=False)
        self.shearangle = ParameterDict(30.0, 0.0, 180.0, fixed=True)

class GaussSource:
    """Stand-in Gaussian source object"""
    def __init__(self):
        self.__class__.__name__ = 'GaussSource'
        self.xoff = ParameterDict(0.1, -0.5, 0.5, fixed=False)
        self.yoff = ParameterDict(0.2, -0.5, 0.5, fixed=False)
        self.flux = ParameterDict(1.0, 0.1, 10.0, fixed=True)
        self.width = ParameterDict(0.05, 0.01, 0.5, fixed=False)

class SersicSource:
    """Stand-in Sersic source object"""
    def __init__(self):
        self.__class__.__name__ = 'SersicSource'
        self.xoff = ParameterDict(0.1, -0.5, 0.5, fixed=False)
        self.yoff = ParameterDict(0.2, -0.5, 0.5, fixed=False)
        self.flux = ParameterDict(1.0, 0.1, 10.0, fixed=False)
        self.majax = ParameterDict(0.2, 0.05, 1.0, fixed=True)
        self.index = ParameterDict(2.0, 0.5, 4.0, fixed=True)
        self.axisratio = ParameterDict(0.7, 0.3, 1.0, fixed=True)
        self.PA = ParameterDict(45.0, 0.0, 180.0, fixed=True)

class PointSource:
    """Stand-in point source object"""
    def __init__(self):
        self.__class__.__name__ = 'PointSource'
        self.xoff = ParameterDict(0.1, -0.5, 0.5, fixed=False)
        self.yoff = ParameterDict(0.2, -0.5, 0.5, fixed=True)
        self.flux = ParameterDict(1.0, 0.1, 10.0, fixed=False)

# Build the objects
lens_objects = [SIELens(), ExternalShear()]
source_objects = [GaussSource(), PointSource()]

# p array: parameters in order of iteration
# From SIELens: x (0.6), y (0.7), M (1.2) - e and PA are fixed
# From ExternalShear: shear (0.075) - shearangle is fixed
# From GaussSource: xoff (0.15), yoff (0.25), width (0.1) - flux is fixed
# From PointSource: xoff (0.12), flux (1.5) - yoff is fixed
p = [0.6, 0.7, 1.2, 0.075, 0.15, 0.25, 0.1, 0.12, 1.5, 1.2, 0.3, 0.4]

scaleamp = [1.0, True, False]
shiftphase = [None, True, False]

ARGS = {
    'p': p,
    'lens': lens_objects,
    'source': source_objects,
    'scaleamp': scaleamp,
    'shiftphase': shiftphase,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pol2cart Ran
ran on a synthesized input
jspilker/visilens/visilens/utils.py
code served (permissive licence) · get_code("fffa74f268492710")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct arguments for pol2cart
# The function takes r and theta as arrays or lists of coordinate pairs

# Create sample polar coordinates
r_values = [1.0, 2.0, 3.0, 4.0]
theta_values = [0.0, np.pi/4, np.pi/2, np.pi]

ARGS = {
    "r": r_values,
    "theta": theta_values,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
uvimageslow Ran
ran on a synthesized input
jspilker/visilens/visilens/plot_images.py
code served (permissive licence) · get_code("8fd5d5b31aca8c81")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (64, 64) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import copy

class Visdata:
    """Stand-in Visdata class with required attributes"""
    def __init__(self, u, v, real, imag, sigma, uvdist):
        self.u = u
        self.v = v
        self.real = real
        self.imag = imag
        self.sigma = sigma
        self.uvdist = uvdist

# Create small test visibility data
n_vis = 8
u = np.linspace(-1, 1, n_vis)
v = np.linspace(-0.5, 0.5, n_vis)
real = np.random.randn(n_vis)
imag = np.random.randn(n_vis)
sigma = np.ones(n_vis) * 0.1
uvdist = np.sqrt(u**2 + v**2)

visdata = Visdata(
    u=u,
    v=v,
    real=real,
    imag=imag,
    sigma=sigma,
    uvdist=uvdist
)

ARGS = {
    'visdata': visdata,
    'imsize': 64,
    'pixsize': 1.0,
    'taper': 0.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (64, 64) — shape (64, 64) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SourceProfile Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/calc_likelihood.py
code served (permissive licence) · get_code("b64e5ae768ccd364")
How it ran

no invocation recorded.

TrianglePlot_MCMC Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/triangleplot.py
code served (permissive licence) · get_code("e74f0fb6d2c60583")
How it ran

no invocation recorded.

bin_visibilities Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/class_utils.py
code served (permissive licence) · get_code("6d9e9c0c063acef5")
How it ran

no invocation recorded.

box Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/utils.py
code served (permissive licence) · get_code("45f52d762752536d")
How it ran

no invocation recorded.

calc_vis_lnlike Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/calc_likelihood.py
code served (permissive licence) · get_code("76b6ed6840c04c0b")
How it ran

no invocation recorded.

marginalize_2d Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/triangleplot.py
code served (permissive licence) · get_code("dfbe9c4048b72abf")
How it ran

no invocation recorded.

read_visdata Not yet run
unverified (harvested, no recorded run)
jspilker/visilens/visilens/class_utils.py
code served (permissive licence) · get_code("024a12327c0fef62")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 14 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The South Pole Telescope has discovered one hundred gravitationally lensed, high-redshift, dusty, star-forming galaxies (DSFGs). We present 0.5" resolution 870um Atacama Large Millimeter/submillimeter Array imaging of a sample of 47 DSFGs spanning z=1.9-5.7, and construct gravitational lens models of these sources. Our visibility-based lens modeling incorporates several sources of residual interferometric calibration uncertainty, allowing us to properly account for noise in the observations. At least 70% of the sources are strongly lensed by foreground galaxies (mu_870um > 2), with a median magnification mu_870um = 6.3, extending to mu_870um > 30. We compare the intrinsic size distribution of the strongly lensed sources to a similar number of unlensed DSFGs and find no significant differences in spite of a bias between the magnification and intrinsic source size. This may indicate that the true size distribution of DSFGs is relatively narrow. We use the source sizes to constrain the wavelength at which the dust optical depth is unity and find this wavelength to be correlated with the dust temperature. This correlation leads to discrepancies in dust mass estimates of a factor of 2 compared to estimates using a single value for this wavelength. We investigate the relationship between the [CII] line and the far-infrared luminosity and find that the same correlation between the [CII]L_FIR ratio and Sigma_FIR found for low-redshift star-forming galaxies applies to high-redshift galaxies and extends at least two orders of magnitude higher in Sigma_FIR. This lends further credence to the claim that the compactness of the IR-emitting region is the controlling parameter in establishing the "[CII] deficit."

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1604.05723")
get_code_for_paper("1604.05723")
have("1604.05723")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/1604.05723.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/1604.05723.svg)](https://syntology.ai/paper/1604.05723)

Connect an agent — have() is free.