SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.03023 · 2023

V1T: large-scale mouse V1 response prediction using a Vision Transformer

arXiv · PDF · Open in the Atlas

Code that ran

16 of 17 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 15 have no contract verdict; 1 failure of our instrument; 1 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/2302.03023.json · badge: /api/ran/2302.03023.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
bryanlimy/V1T canonical (classifier label)
classifier basis not recorded on the link
16 ran · 1 no recorded run
FunctionStatusWhere it lives
attention_rollout Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/utils/attention_rollout.py
code served (permissive licence) · get_code("25d91072d17ab666")
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 (224, 224) 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 math

torch.manual_seed(0)

# The function expects:
# - attention: torch.Tensor with dim() == 4, shape should be (num_layers, num_heads, seq_len, seq_len)
# - image_shape: List[int] for the target image dimensions

# Create a 4D attention tensor
# Shape: (num_layers=2, num_heads=3, seq_len=5, seq_len=5)
# This represents attention maps from a vision transformer with 2 layers, 3 heads, 5 token positions
attention = torch.randn(2, 3, 5, 5)
attention = torch.softmax(attention, dim=-1)  # Make it valid softmax attention

# image_shape should be a list of integers representing target height and width
image_shape = [224, 224]

ARGS = {
    "attention": attention,
    "image_shape": image_shape,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — 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 (224, 224) — shape (224, 224) 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
conv2d_shape Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/models/utils.py
code served (permissive licence) · get_code("a3e5931b36cec32e")
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 1, 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 torch
import numpy as np
from math import floor

torch.manual_seed(0)

# Test case 1: Basic conv2d shape calculation
# Input shape is (C, H, W) = (3, 32, 32) - 3 channels, 32x32 spatial
# num_filters = 16
# kernel_size = 3x3
# stride = 1
# padding = 1
# dilation = 1

ARGS = {
    "input_shape": (3, 32, 32),
    "num_filters": 16,
    "kernel_size": 3,
    "stride": 1,
    "padding": 1,
    "dilation": 1
}
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 tuple, ndim 1, 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
find_shape Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/utils/attention_rollout.py
code served (permissive licence) · get_code("d386906a19182d5f")
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 1, 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 math

def find_shape(num_patches: int):
    dim1 = math.ceil(math.sqrt(num_patches))
    while num_patches % dim1 != 0 and dim1 > 0:
        dim1 -= 1
    dim2 = num_patches // dim1
    return (dim1, dim2)

ARGS = {
    "num_patches": 12
}
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, ndim 1, 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
get_image_shape Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/data.py
code served (permissive licence) · get_code("8c76c67923e7047e")
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 1, 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 os
import numpy as np
import tempfile
import shutil

# Create a temporary directory structure to simulate the mouse_dir layout
temp_dir = tempfile.mkdtemp()
data_dir = os.path.join(temp_dir, 'data')
images_dir = os.path.join(data_dir, 'images')
os.makedirs(images_dir, exist_ok=True)

# Create a dummy image file
dummy_image = np.random.rand(32, 64, 3).astype(np.float32)
np.save(os.path.join(images_dir, '0.npy'), dummy_image)

ARGS = {
    'mouse_dir': temp_dir
}
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 1, 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
get_mouse2path Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/data.py
code served (permissive licence) · get_code("720b97c4c213f436")
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 torch
import typing as t

# Define the function to test
DS_NAMES = t.Literal['sensorium', 'franke2022']

SENSORIUM = {'S0': 'static26872-17-20-GrayImageNet-94c6ff995dac583098847cfecd43e7b6', 'S1': 'static27204-5-13-GrayImageNet-94c6ff995dac583098847cfecd43e7b6', 'A': 'static21067-10-18-GrayImageNet-94c6ff995dac583098847cfecd43e7b6', 'B': 'static22846-10-16-GrayImageNet-94c6ff995dac583098847cfecd43e7b6', 'C': 'static23343-5-17-GrayImageNet-94c6ff995dac583098847cfecd43e7b6', 'D': 'static23656-14-22-GrayImageNet-94c6ff995dac583098847cfecd43e7b6', 'E': 'static23964-4-22-GrayImageNet-94c6ff995dac583098847cfecd43e7b6'}

FRANKE2022 = {'F': 'static25311-10-26-ColorImageNet-104e446ed0128d89c639eef0abe4655b', 'G': 'static25340-3-19-ColorImageNet-104e446ed0128d89c639eef0abe4655b', 'H': 'static25704-2-12-ColorImageNet-b23ac8521543becfd382e56c657ba29b', 'I': 'static25830-10-4-ColorImageNet-104e446ed0128d89c639eef0abe4655b', 'J': 'static26085-6-3-ColorImageNet-104e446ed0128d89c639eef0abe4655b', 'K': 'static26142-2-11-ColorImageNet-6a21297215f4dbb802554a60c0e72877', 'L': 'static26426-18-13-ColorImageNet-b23ac8521543becfd382e56c657ba29b', 'M': 'static26470-4-5-ColorImageNet-104e446ed0128d89c639eef0abe4655b', 'N': 'static26644-6-2-ColorImageNet-b23ac8521543becfd382e56c657ba29b', 'O': 'static26872-21-6-ColorImageNet-104e446ed0128d89c639eef0abe4655b'}

def get_mouse2path(ds_name: DS_NAMES):
    assert ds_name in ('sensorium', 'franke2022')
    return SENSORIUM if ds_name == 'sensorium' else FRANKE2022

# Create setup arguments
ARGS = {
    'ds_name': 'sensorium'
}
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 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
get_num_trials Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/data.py
code served (permissive licence) · get_code("c04bd02a49089908")
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 int, ndim 0, dtype family int, 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 os
import tempfile
import numpy as np
from glob import glob

# Create a temporary directory structure that mimics a mouse_dir
temp_dir = tempfile.mkdtemp()
mouse_dir = temp_dir
data_dir = os.path.join(mouse_dir, 'data', 'images')
os.makedirs(data_dir, exist_ok=True)

# Create some dummy .npy files
for i in range(5):
    np.save(os.path.join(data_dir, f'trial_{i}.npy'), np.array([1, 2, 3]))

ARGS = {
    'mouse_dir': mouse_dir
}
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 int, ndim 0, dtype family int, 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
int2tuple Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/models/utils.py
code served (permissive licence) · get_code("13efeb513bfdf3ae")
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 1, 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 typing as t

def int2tuple(value: t.Union[int, t.Tuple[int, int]]):
    return (value, value) if isinstance(value, int) else value

# Test case 1: integer input
ARGS = {
    "value": 5
}
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, ndim 1, 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
msse Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/losses.py
code served (permissive licence) · get_code("3383fa71e8022690")
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
import typing as t

torch.manual_seed(0)

# Create small tensors for y_true and y_pred
# Using batch size 2 and feature dimension 4
y_true = torch.randn(2, 4)
y_pred = torch.randn(2, 4)

# reduction parameter with default value 'sum'
reduction = 'sum'

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
    'reduction': reduction
}
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
normalize Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/utils/attention_rollout.py
code served (permissive licence) · get_code("8549754f0b67b989")
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)

# The function normalize takes a single parameter x which can be either
# a numpy array or a torch tensor. It performs min-max normalization.
# We'll test with both types.

# Create a small numpy array
numpy_array = np.array([[1.0, 2.0, 3.0, 4.0],
                        [5.0, 6.0, 7.0, 8.0]])

# Create a small torch tensor
torch_tensor = torch.tensor([[1.0, 2.0, 3.0, 4.0],
                             [5.0, 6.0, 7.0, 8.0]])

# We'll use the numpy array as the primary test case
ARGS = {
    "x": numpy_array
}
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
num_correct Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/pretrain/classification.py
code served (permissive licence) · get_code("945936d5a20207ce")
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 float, 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 small tensors for y_true and y_pred
# y_true: ground truth labels
# y_pred: predicted labels
# Both should have the same shape for comparison
y_true = torch.tensor([0, 1, 2, 1])
y_pred = torch.tensor([0, 1, 2, 0])

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
}
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 float, 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
poisson_loss Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/losses.py
code served (permissive licence) · get_code("ffee22053b603bfe")
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 small tensors for y_true and y_pred
# y_true: true counts (should be non-negative)
# y_pred: predicted values (should be positive for log stability)
y_true = torch.tensor([[1.0, 2.0, 3.0, 0.0],
                       [2.0, 1.0, 0.0, 1.0]], dtype=torch.float32)

y_pred = torch.tensor([[1.1, 2.1, 2.9, 0.1],
                       [2.2, 0.9, 0.2, 1.1]], dtype=torch.float32)

# eps: small value for numerical stability
eps = 1e-12

# reduction: one of 'sum' or 'mean'
reduction = 'sum'

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
    'eps': eps,
    'reduction': reduction
}
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
pool2d_shape Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/models/utils.py
code served (permissive licence) · get_code("f8da76b35279fe40")
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 1, 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 torch
from math import floor
import typing as t

torch.manual_seed(0)

ARGS = {
    "input_shape": (2, 16, 20),
    "kernel_size": 3,
    "stride": 2,
    "padding": 1,
    "dilation": 1
}
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 tuple, ndim 1, 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
register Ran
ran; our contract draft was wrong, not the code
bryanlimy/V1T/src/v1t/losses.py
code served (permissive licence) · get_code("3c702d811d125a17")
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 function. 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 torch
import torch.nn as nn

torch.manual_seed(0)

# The register function takes one parameter: name (a string)
# It returns a decorator function that adds functions to the _CRITERION dict
# To test this, we need to call register with a name argument

name = "loss_function"

ARGS = {
    "name": name
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
reverse Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/pretrain/data.py
code served (permissive licence) · get_code("162fbc498c065ad9")
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 4, dtype family float, shape (2, 3, 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 torch

torch.manual_seed(0)

# The function takes a standardized image tensor and reverses the standardization
# by multiplying by IMAGE_STD and adding IMAGE_MEAN.
# We need to create a small standardized image tensor.

# Create a small standardized image tensor (batch_size=2, channels=3, height=4, width=4)
standardized_image = torch.randn(2, 3, 4, 4)

ARGS = {
    "image": standardized_image
}
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 4, dtype family float, shape (2, 3, 4, 4) — shape (2, 3, 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
ssim Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/pretrain/reconstruction.py
code served (permissive licence) · get_code("8cb6d2623964ee80")
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
import typing as t

torch.manual_seed(0)

# Create small 4D tensors: (batch=2, channels=3, height=4, width=5)
x = torch.randn(2, 3, 4, 5)
y = torch.randn(2, 3, 4, 5)

ARGS = {
    'x': x,
    'y': y,
    'max_value': 255.0,
    'win_size': 11,
    'win_sigma': 1.5,
    'K1': 0.01,
    'K2': 0.03,
    'reduction': 'mean'
}
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
transform Ran
ran on a synthesized input
bryanlimy/V1T/src/v1t/pretrain/data.py
code served (permissive licence) · get_code("9b4e97266208c605")
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 (1, 36, 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 torch
from PIL import Image
import numpy as np

torch.manual_seed(0)

# Create a small test image
# PIL Image needs to be in RGB or grayscale mode
# Create a simple RGB image (3 channels, 100x100)
img_array = np.random.randint(0, 256, (100, 100, 3), dtype=np.uint8)
test_image = Image.fromarray(img_array, mode='RGB')

# Create arguments dict
ARGS = {
    'image': test_image,
    'resize_image': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — 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 (1, 36, 64) — shape (1, 36, 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
get_ds Not yet run
unverified (harvested, no recorded run)
bryanlimy/V1T/src/v1t/pretrain/data.py
code served (permissive licence) · get_code("7b7d4b5661b63a5a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 16 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
register 1 other paper
1 later
2407.09100 (2024-07)
Retrospective for the Dynamic Sensorium Competition for predicting large-scale mouse primary visual cortex activity from videos

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

Accurate predictive models of the visual cortex neural response to natural visual stimuli remain a challenge in computational neuroscience. In this work, we introduce V1T, a novel Vision Transformer based architecture that learns a shared visual and behavioral representation across animals. We evaluate our model on two large datasets recorded from mouse primary visual cortex and outperform previous convolution-based models by more than 12.7% in prediction performance. Moreover, we show that the self-attention weights learned by the Transformer correlate with the population receptive fields. Our model thus sets a new benchmark for neural response prediction and can be used jointly with behavioral and neural recordings to reveal meaningful characteristic features of the visual cortex.

For agents

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

get_harvested_code_for_paper("2302.03023")
get_code_for_paper("2302.03023")
have("2302.03023")

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

curl https://syntology.ai/api/ran/2302.03023.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.03023.svg)](https://syntology.ai/paper/2302.03023)

Connect an agent — have() is free.