SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2301.11955 · 2023

Adaptive whitening in neural populations with gain-modulating interneurons

arXiv · PDF · Open in the Atlas

Code that ran

19 of 19 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 19 have no contract verdict; 0 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/2301.11955.json · badge: /api/ran/2301.11955.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
lyndond/frame_whitening canonical (classifier label)
classifier basis not recorded on the link
19 ran · 0 no recorded run
FunctionStatusWhere it lives
adapt_covariance Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/simulation.py
code served (permissive licence) · get_code("a16362a9b54e9c9d")
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
from collections.abc import Sequence

# Set seed for reproducibility
np.random.seed(42)

# Create small test covariance matrices
# N=4 (dimension), will create 2 contexts
N = 4
K = 3  # frame dimension

# Create positive definite covariance matrices
# First context
A1 = np.random.randn(N, N)
Cxx1 = A1 @ A1.T + np.eye(N)

# Second context
A2 = np.random.randn(N, N)
Cxx2 = A2 @ A2.T + np.eye(N)

Cxx_list = [Cxx1, Cxx2]

# Create frame W (N x K)
W = np.random.randn(N, K)

# Initial gains
g0 = np.ones(K)

# Define all arguments
ARGS = {
    'Cxx_list': Cxx_list,
    'W': W,
    'batch_size': 2,
    'n_batch': 4,
    'lr_g': 0.001,
    'g0': g0,
    'online': True,
    'clamp': False,
    'alpha': 1.0,
    'save_every': 1,
    'seed': 42,
    'verbose': False,
    'break_on_convergence': False,
    'break_on_divergence': True,
    'error_type': 'fro'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tqdm — 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
compute_error Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/simulation.py
code served (permissive licence) · get_code("1ccd9435f7664f94")
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 float64, ndim 0, 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 a small positive semi-definite covariance matrix
np.random.seed(0)
N = 4
# Generate a random matrix and make it symmetric positive semi-definite
A = np.random.randn(N, N).astype(np.float64)
Cxx = (A @ A.T).astype(np.float64)

# Create a transformation matrix M
M = np.random.randn(N, N).astype(np.float64)

ARGS = {
    'M': M,
    'Cxx': Cxx,
    'clamp': True,
    'error_type': 'fro'
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
get_benucci_data Ran
ran on a synthesized input
lyndond/frame_whitening/figures/neural_data/benucci_carandini.py
code served (permissive licence) · get_code("4f889521fb32e2e7")
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
from pathlib import Path
from scipy.io import savemat
import tempfile

# Create a temporary directory and a mock .mat file
temp_dir = tempfile.mkdtemp()
temp_data_path = Path(temp_dir) / 'BenucciTuningData.mat'

# Create mock data structure matching what the function expects
# The function accesses data_v1['resps'][i][0][..., 0] and data_v1['resps'][i][0][..., 1]
# So 'resps' should be a list/array of objects, each with shape (1, n_orientations, 2)

resps_data = []
for i in range(11):  # 11 probability conditions
    # Each element: shape (1, 21, 2) - the middle dimension is orientations (21 values from -90 to 90)
    # Last dimension has unadapted (0) and adapted (1) responses
    condition_data = np.random.randn(1, 21, 2)
    resps_data.append(condition_data)

# Save as .mat file
savemat(str(temp_data_path), {'resps': np.array(resps_data, dtype=object)})

ARGS = {
    'data_path': temp_data_path,
    'subset': 'all'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
get_benucci_prob Ran
ran on a synthesized input
lyndond/frame_whitening/figures/neural_data/benucci_carandini.py
code served (permissive licence) · get_code("25799dd05df935ac")
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 1, dtype family float, shape (21,) 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

# Define arguments for get_benucci_prob
ARGS = {
    "prob_adaptor": 0.7,
    "n_stim_bins": 21,
    "peak": 0.5
}
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 1, dtype family float, shape (21,) — shape (21,) 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
get_g_opt Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/simulation.py
code served (permissive licence) · get_code("bd3bc868b78011f2")
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 1, dtype family float, shape (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 numpy as np
import scipy as sp

np.random.seed(0)

# W is shape (N, K) - used in matrix operations
# Cxx is shape (N, N) - used in matrix square root and operations
N = 5
K = 3

# Create W: shape (N, K)
W = np.random.randn(N, K)

# Create Cxx: shape (N, N) - symmetric positive definite
# Generate as A @ A.T to ensure positive definiteness
A = np.random.randn(N, N)
Cxx = A @ A.T

ARGS = {
    "W": W,
    "Cxx": Cxx,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (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
get_mercedes_frame Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/frame.py
code served (permissive licence) · get_code("7915df461e5c3fd6")
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, 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 numpy as np

# Define arguments for get_mercedes_frame
ARGS = {
    "parseval": False,
    "jitter": False,
}
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 (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
get_near_etf Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/_deprecated/_frame.py
code served (permissive licence) · get_code("7e5085acb1a94478")
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 (5, 10) 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

# Set random seed for reproducibility
np.random.seed(0)

ARGS = {
    "n": 5,
    "k": 10,
    "s": None,
    "thresh": 0.5,
    "tight_iter": 0,
    "num_iter": None,
    "show": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, tqdm — 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 (5, 10) — shape (5, 10) 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
get_patches Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/data.py
code served (permissive licence) · get_code("c80f18acf5286ac0")
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 list, ndim 3, dtype family int; 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 a small image array (height=10, width=8)
img = np.random.randint(0, 256, size=(10, 8), dtype=np.uint8)

# Define patch dimensions
patch_h = 3
patch_w = 4

# Define walk arrays - offsets for extracting patches
# These represent starting positions (di, dj) for each patch
walk_h = np.array([0, 2, 5], dtype=np.int64)
walk_w = np.array([0, 2, 4], dtype=np.int64)

ARGS = {
    "img": img,
    "patch_h": patch_h,
    "patch_w": patch_w,
    "walk_h": walk_h,
    "walk_w": walk_w,
}
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 list, ndim 3, dtype family int — a list 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
load_images Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/data.py
code served (permissive licence) · get_code("0cc0051da36e0a0c")
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 list, ndim 3, dtype family int; 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 array
import os
import os.path as op
import tempfile
import numpy as np

# Create a temporary directory with mock image files
temp_dir = tempfile.mkdtemp()

# Create mock image files in van hateren format (binary uint16 big-endian)
# van hateren images are 1024x1536 = 1,572,864 pixels
for i in range(3):
    filename = op.join(temp_dir, f'image_{i:05d}.raw')
    # Create a small mock image (1024x1536 uint16)
    mock_data = np.random.randint(0, 256, size=(1024, 1536), dtype=np.uint16)
    # Convert to big-endian bytes (as van hateren format requires)
    mock_data_be = mock_data.astype('>u2')
    with open(filename, 'wb') as f:
        f.write(mock_data_be.tobytes())

# Set up arguments
ARGS = {
    'data_dir': temp_dir,
    'n_images': 2,
    'crop_size': 256,
    'rng': np.random.default_rng(seed=42)
}
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 list, ndim 3, dtype family int — a list 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
load_matfile Ran
ran on a synthesized input
lyndond/frame_whitening/figures/neural_data/wanner_friedrich.py
code served (permissive licence) · get_code("c0ea292253a9d2aa")
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 tempfile
import scipy.io as scio
import numpy as np

# Create a temporary .mat file with some test data
temp_dir = tempfile.gettempdir()
temp_filename = f"{temp_dir}/test_matfile.mat"

# Create sample data to save
test_data = {
    'simple_array': np.array([1, 2, 3, 4, 5]),
    'nested_object': np.array([np.array([1, 2]), np.array([3, 4])], dtype=object),
    '_internal_key': np.array([999, 999]),  # Should be skipped
}

# Save to .mat file
scio.savemat(temp_filename, test_data)

ARGS = {
    'filename': temp_filename
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
normalize_cov Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/stats.py
code served (permissive licence) · get_code("805f5522caf910ea")
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 (4, 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 numpy as np

# Create a small covariance matrix for testing
# A valid covariance matrix should be symmetric and positive semi-definite
np.random.seed(0)
size = 4
# Create a random matrix and make it symmetric positive semi-definite
A = np.random.randn(size, size)
C = A @ A.T  # This ensures symmetric and positive semi-definite
C = C.astype(np.float64)

ARGS = {
    "C": C
}
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 (4, 4) — shape (4, 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
normalize_frame Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/frame.py
code served (permissive licence) · get_code("87dc54f865cb861b")
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 (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 numpy as np

# Create a small test matrix with distinct dimensions
# 4 rows, 3 columns - will normalize along axis 0 (columns) by default
W = np.array([
    [1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0],
    [7.0, 8.0, 9.0],
    [10.0, 11.0, 12.0]
], dtype=np.float64)

ARGS = {
    "W": W,
    "axis": 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 (4, 3) — shape (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
normalize_pdf Ran
ran on a synthesized input
lyndond/frame_whitening/figures/neural_data/benucci_carandini.py
code served (permissive licence) · get_code("672dd3a8b5370a76")
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 1, dtype family float, shape (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 numpy as np

# Create a simple 1D probability distribution (flat vector)
p_flat = np.array([0.2, 0.3, 0.1, 0.4], dtype=np.float64)

ARGS = {
    "p": p_flat
}
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 1, dtype family float, shape (4,) — shape (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
orthogonal2 Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/_deprecated/_frame.py
code served (permissive licence) · get_code("92571b220f636988")
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, 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 numpy as np

ARGS = {
    "th": 0.5
}
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 (2, 2) — shape (2, 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
parsevalize Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/frame.py
code served (permissive licence) · get_code("dabd4280bd85359f")
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 (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 numpy as np
from scipy.linalg import fractional_matrix_power

# Create a small frame matrix W
# W should be shape (n, m) where n >= m for a valid frame
# Let's use a 4x3 matrix (more rows than columns)
np.random.seed(0)
W = np.random.randn(4, 3).astype(np.float64)

ARGS = {
    "W": W
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 (4, 3) — shape (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
process_cell_info Ran
ran on a synthesized input
lyndond/frame_whitening/figures/neural_data/wanner_friedrich.py
code served (permissive licence) · get_code("b9378e10401c82dd")
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 DataFrame. 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 pandas as pd

# Construct a valid cell_info_array
# The function expects a numpy array with 6 columns:
# Column 0: ID (will be cast to int)
# Column 1: row_idx (will be cast to int, then -1)
# Column 2: cell_type (will be cast to int)
# Column 3: microcluster (will be cast to int)
# Column 4: GlomID (will be cast to int)
# Column 5: length (will be cast to float)

cell_info_array = np.array([
    [1, 2, 10, 5, 100, 3.14],
    [2, 3, 11, 6, 101, 2.71],
], dtype=object)

ARGS = {
    'cell_info_array': cell_info_array
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
randcov2 Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/stats.py
code served (permissive licence) · get_code("7ae8b761c0914299")
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

ARGS = {
    "cond_max": 3.0,
    "l_max": 2.0
}
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
randcovn Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/stats.py
code served (permissive licence) · get_code("dcf21358bb11470f")
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

ARGS = {
    "N": 5,
    "cond_max": 3.0,
    "l_max": 2.0
}
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
random_walk Ran
ran on a synthesized input
lyndond/frame_whitening/frame_whitening/data.py
code served (permissive licence) · get_code("62916b3f382d9728")
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 int; 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
import numpy.typing as npt
from typing import Optional, Tuple

# Set random seed for reproducibility
np.random.seed(42)

# Create a random number generator
rng = np.random.default_rng(seed=42)

# Define ARGS dict with all parameters
ARGS = {
    "n_steps": 10,
    "sigma": 1.5,
    "rng": rng
}
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 int — 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

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 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 19 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

Statistical whitening transformations play a fundamental role in many computational systems, and may also play an important role in biological sensory systems. Existing neural circuit models of adaptive whitening operate by modifying synaptic interactions; however, such modifications would seem both too slow and insufficiently reversible. Motivated by the extensive neuroscience literature on gain modulation, we propose an alternative model that adaptively whitens its responses by modulating the gains of individual neurons. Starting from a novel whitening objective, we derive an online algorithm that whitens its outputs by adjusting the marginal variances of an overcomplete set of projections. We map the algorithm onto a recurrent neural network with fixed synaptic weights and gain-modulating interneurons. We demonstrate numerically that sign-constraining the gains improves robustness of the network to ill-conditioned inputs, and a generalization of the circuit achieves a form of local whitening in convolutional populations, such as those found throughout the visual or auditory systems.

For agents

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

get_harvested_code_for_paper("2301.11955")
get_code_for_paper("2301.11955")
have("2301.11955")

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

curl https://syntology.ai/api/ran/2301.11955.json

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

[![Syntology run record](https://syntology.ai/api/ran/2301.11955.svg)](https://syntology.ai/paper/2301.11955)

Connect an agent — have() is free.