SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1910.13110 · 2019

Unsupervised Deep Basis Pursuit: Learning inverse problems without ground-truth data

arXiv · PDF · Open in the Atlas

Code that ran

11 of 11 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 9 have no contract verdict; 2 failures of our instrument; 0 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/1910.13110.json · badge: /api/ran/1910.13110.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
utcsilab/deepinpy canonical (classifier label)
classifier basis: census
11 ran · 0 no recorded run
FunctionStatusWhere it lives
dot Ran
ran; our contract draft was wrong, not the code
utcsilab/deepinpy/deepinpy/opt/opt.py
code served (permissive licence) · get_code("ea74f30040383586")
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 Tensor, ndim 0, dtype family float, shape () 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 torch

torch.manual_seed(0)

x1 = torch.randn(4)
x2 = torch.randn(4)

ARGS = {
    "x1": x1,
    "x2": x2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () 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
dot_batch Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/opt/opt.py
code served (permissive licence) · get_code("309667508c01a3e1")
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 Tensor, ndim 1, dtype family float, shape (2,) 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 torch

torch.manual_seed(0)

# Create two small multidimensional tensors with batch dimension
# x1: batch_size=2, feature_dim=4
x1 = torch.randn(2, 4)

# x2: batch_size=2, feature_dim=4 (must match x1's shape for element-wise multiplication)
x2 = torch.randn(2, 4)

ARGS = {
    "x1": x1,
    "x2": x2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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
dot_single Ran
ran; our contract draft was wrong, not the code
utcsilab/deepinpy/deepinpy/opt/opt.py
code served (permissive licence) · get_code("ad0c8005480957cb")
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 Tensor, ndim 0, dtype family float, shape () 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 torch

torch.manual_seed(0)

# Create a small vector tensor for dot_single
x = torch.randn(4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () 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
fft_forw Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/forwards/mcmri/mcmri.py
code served (permissive licence) · get_code("df41a873dc2d9b11")
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 Tensor, ndim 3, dtype family float, shape (2, 4, 3) 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 torch

torch.manual_seed(0)

# fft_forw takes parameters: x and ndim
# x should be a tensor that torch.fft.fftn can process
# ndim should be an integer (2 or 3 based on the function logic)

# Create a small complex tensor for FFT input
# torch.fft.fftn expects a real or complex tensor
# Shape: (2, 4, 3) - batch of 2, with spatial dimensions 4 and 3
x = torch.randn(2, 4, 3, dtype=torch.float32)

ARGS = {
    'x': x,
    'ndim': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 3) 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
h5_read Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/utils/utils.py
code served (permissive licence) · get_code("3c0c5d9b68295c24")
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 dict. 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
import h5py
import tempfile
import os

# Create a temporary HDF5 file with some test data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_data.h5")

with h5py.File(temp_file, 'w') as f:
    f.create_dataset('array_1', data=np.array([1, 2, 3, 4]))
    f.create_dataset('array_2', data=np.array([[1.0, 2.0], [3.0, 4.0]]))
    f.create_dataset('array_3', data=np.random.randn(2, 3, 4))

ARGS = {
    'filename': temp_file,
    'key_list': ['array_1', 'array_2', 'array_3', 'nonexistent_key']
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
load_data Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/forwards/mcmri/dataset.py
code served (permissive licence) · get_code("76b9cd95667914d2")
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
import h5py
import tempfile
import os

# Create a temporary HDF5 file with the required datasets
temp_dir = tempfile.mkdtemp()
data_file = os.path.join(temp_dir, "test_data.h5")

with h5py.File(data_file, 'w') as f:
    # Create datasets with distinct dimensions for probing
    # imgs: shape (10, 2, 4, 3) - 10 samples, 2 coils, 4x3 spatial
    f.create_dataset('imgs', data=np.random.randn(10, 2, 4, 3).astype(np.complex64))
    
    # maps: shape (10, 2, 4, 3) - sensitivity maps same shape as imgs
    f.create_dataset('maps', data=np.random.randn(10, 2, 4, 3).astype(np.complex64))
    
    # masks: shape (10, 4, 3) - undersampling mask (2D spatial)
    f.create_dataset('masks', data=np.random.rand(10, 4, 3).astype(np.float32))
    
    # noise: shape (10, 2, 4, 3) - optional noise
    f.create_dataset('noise', data=np.random.randn(10, 2, 4, 3).astype(np.complex64))

ARGS = {
    'idx': 0,
    'data_file': data_file,
    'gen_masks': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, 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
load_data_ksp Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/forwards/mcmri/dataset.py
code served (permissive licence) · get_code("3e33abe79d67a2b3")
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
import h5py
import tempfile
import os

# Create a temporary HDF5 file with the required datasets
temp_dir = tempfile.mkdtemp()
data_file = os.path.join(temp_dir, 'test_data.h5')

with h5py.File(data_file, 'w') as f:
    # Create datasets with appropriate shapes and dtypes
    # Shape: (num_samples, height, width, coils) for ksp/maps
    # Shape: (num_samples, 1, width) for masks
    # Shape: (num_samples, height, width) for imgs
    
    num_samples = 3
    height = 4
    width = 5
    coils = 2
    
    # imgs: complex64, shape (num_samples, height, width)
    imgs_data = np.random.randn(num_samples, height, width).astype(np.complex64)
    f.create_dataset('imgs', data=imgs_data)
    
    # maps: complex64, shape (num_samples, height, width, coils)
    maps_data = np.random.randn(num_samples, height, width, coils).astype(np.complex64)
    f.create_dataset('maps', data=maps_data)
    
    # ksp: complex64, shape (num_samples, height, width, coils)
    ksp_data = np.random.randn(num_samples, height, width, coils).astype(np.complex64)
    f.create_dataset('ksp', data=ksp_data)
    
    # masks: float32, shape (num_samples, 1, width)
    masks_data = np.random.rand(num_samples, 1, width).astype(np.float32)
    f.create_dataset('masks', data=masks_data)
    
    # noise: float32, shape (num_samples, height, width)
    noise_data = np.random.rand(num_samples, height, width).astype(np.float32)
    f.create_dataset('noise', data=noise_data)

ARGS = {
    'idx': 0,
    'data_file': data_file,
    'gen_masks': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, 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
maps_adj Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/forwards/mcmri/mcmri.py
code served (permissive licence) · get_code("d93fc0f2daad1714")
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 Tensor, ndim 2, dtype family float, shape (2, 3) 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 torch

torch.manual_seed(0)

# cimg: tensor that will be element-wise multiplied with conj(maps) and summed
# maps: tensor that will be conjugated and multiplied with cimg
# The function does: torch.sum(torch.conj(maps) * cimg, 1, keepdim=False)
# So both tensors need compatible shapes for broadcasting/multiplication
# Shape analysis: sum over dimension 1, so both should have at least 2 dimensions
# Let's use shape (batch=2, channels=4, spatial=3) for maps
# and shape (batch=2, channels=4, spatial=3) for cimg

cimg = torch.randn(2, 4, 3)
maps = torch.randn(2, 4, 3, dtype=torch.complex64)

ARGS = {
    "cimg": cimg,
    "maps": maps,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) 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
maps_forw Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/forwards/mcmri/mcmri.py
code served (permissive licence) · get_code("46df83120af42e0c")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 4) 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Analyze the function:
# - img: indexed with [:, None, ...] so it's a tensor
# - maps: multiplied with img[:, None, ...], so it's a tensor
# - The operation broadcasts img (adding a dimension at position 1) with maps

# Create img tensor: batch=2, other dims=4
img = torch.randn(2, 4)

# Create maps tensor: must be broadcastable with img[:, None, ...]
# img[:, None, ...] has shape (2, 1, 4)
# maps should have shape compatible for broadcasting
maps = torch.randn(2, 3, 4)

ARGS = {
    "img": img,
    "maps": maps,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 4) 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
t2n2 Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/utils/utils.py
code served (permissive licence) · get_code("9b4a91f1da276ade")
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 (2, 4) 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 torch
import numpy as np

torch.manual_seed(0)

# Create a small tensor to pass to t2n2
x = torch.randn(2, 4)

ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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 (2, 4) — shape (2, 4) 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
topk Ran
ran on a synthesized input
utcsilab/deepinpy/deepinpy/utils/utils.py
code served (permissive licence) · get_code("512aa16f399b3b19")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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 torch

torch.manual_seed(0)

# Create a small tensor with shape (2, 4) to use as input
inp = torch.randn(2, 4)

# Define k as the number of top values to retrieve
k = 2

# Define dim as the dimension along which to retrieve top k values
dim = 1

ARGS = {
    "inp": inp,
    "k": k,
    "dim": dim
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 9 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.

FunctionAlso present inEarliest-datedOthers named
dot 1 other paper
1 later
2211.00199 (2022-11)
Accelerated Motion Correction with Deep Generative Diffusion Models
dot_single 1 other paper
1 later
2211.00199 (2022-11)
Accelerated Motion Correction with Deep Generative Diffusion Models

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

Basis pursuit is a compressed sensing optimization in which the l1-norm is minimized subject to model error constraints. Here we use a deep neural network prior instead of l1-regularization. Using known noise statistics, we jointly learn the prior and reconstruct images without access to ground-truth data. During training, we use alternating minimization across an unrolled iterative network and jointly solve for the neural network weights and training set image reconstructions. At inference, we fix the weights and pass the measurements through the network. We compare reconstruction performance between unsupervised and supervised (i.e. with ground-truth) methods. We hypothesize this technique could be used to learn reconstruction when ground-truth data are unavailable, such as in high-resolution dynamic MRI.

For agents

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

get_harvested_code_for_paper("1910.13110")
get_code_for_paper("1910.13110")
have("1910.13110")

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

curl https://syntology.ai/api/ran/1910.13110.json

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

[![Syntology run record](https://syntology.ai/api/ran/1910.13110.svg)](https://syntology.ai/paper/1910.13110)

Connect an agent — have() is free.