SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.05916 · NeurIPS · 2023

SwiFT: Swin 4D fMRI Transformer

Taesup Moon, Peter Kim, Junbeom Kwon, Sunghwan Joo, Sangyoon Bae, Donggyu Lee, Yoonho Jung, Shinjae Yoo, Jiook Cha

arXiv · PDF · Open in the Atlas

Code that ran

25 of 42 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 19 failures of our instrument; 17 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2307.05916.json · badge: /api/ran/2307.05916.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
Project-MONAI/MONAI canonical (classifier label)
classifier basis: census
9 ran · 0 no recorded run
transconnectome/swift canonical (classifier label)
classifier basis: classifier_2025
8 ran · 4 no recorded run
Wayfear/BrainNetworkTransformer canonical (classifier label)
classifier basis: census
7 ran · 3 no recorded run
DAVEISHAN/TCLR community
Papers with Code archive
1 ran · 3 no recorded run
GonyRosenman/TFF community
Papers with Code archive
0 ran · 7 no recorded run
FunctionStatusWhere it lives
window_partition Ran
ran; our contract draft was wrong, not the code
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("e0d194741047c322")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 3, dtype family float, shape (96, 40, 3) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-05T03:08:04Z). 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 6D tensor with shape (B, D, H, W, T, C)
# B=2 (batch), D=4 (depth), H=6 (height), W=8 (width), T=10 (time), C=3 (channels)
# Dimensions chosen to be divisible by window_size and distinct from each other
x = torch.randn(2, 4, 6, 8, 10, 3)

# window_size is a tuple/list of 4 elements for [D, H, W, T] dimensions
window_size = [2, 2, 2, 5]

ARGS = {
    "x": x,
    "window_size": window_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 (96, 40, 3) — shape (96, 40, 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
PatchEmbed Ran
ran on a synthesized input · constructed an object; no result was computed
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("6c813e3ad36d54c6")
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 an instance of its own class, PatchEmbed: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Create a simple norm layer (LayerNorm)
norm_layer = nn.LayerNorm

# Define ARGS dict with all parameters for PatchEmbed
ARGS = {
    'img_size': (96, 96, 96, 20),
    'patch_size': (4, 4, 4, 1),
    'in_chans': 2,
    'embed_dim': 24,
    'norm_layer': norm_layer,
    'flatten': True,
    'spatial_dims': 3
}
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 PatchEmbed — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PatchEmbed 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
PatchMergingV2 Ran
ran on a synthesized input · constructed an object; no result was computed
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("328b4af4918629c8")
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 an instance of its own class, PatchMergingV2: the run constructed the object and did not call it, so no result was computed. 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
from torch.nn import LayerNorm

torch.manual_seed(0)

ARGS = {
    "dim": 4,
    "norm_layer": nn.LayerNorm,
    "spatial_dims": 3,
    "c_multiplier": 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 PatchMergingV2 — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PatchMergingV2 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
PositionalEmbedding Ran
ran on a synthesized input · constructed an object; no result was computed
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("68b986bf66c51ea9")
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 an instance of its own class, PositionalEmbedding: the run constructed the object and did not call it, so no result was computed. 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
from monai.networks.layers import DropPath, trunc_normal_

torch.manual_seed(0)

ARGS = {
    "dim": 4,
    "patch_dim": (2, 3, 4, 5),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
monai, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PositionalEmbedding — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PositionalEmbedding 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
SpatioTemporalConv Ran
ran on a synthesized input
DAVEISHAN/TCLR/tclr_pretraining/r3d_prp.py
code served (permissive licence) · get_code("2c09aa35ea6cf969")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 5, dtype family float, shape (2, 4, 8, 16, 16) 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
from torch.nn.modules.utils import _triple

class SpatioTemporalConv(nn.Module):
    """Applies a factored 3D convolution over an input signal composed of several input
    planes with distinct spatial and time axes, by performing a 2D convolution over the
    spatial axes to an intermediate subspace, followed by a 1D convolution over the time
    axis to produce the final output.
    Args:
        in_channels (int): Number of channels in the input tensor
        out_channels (int): Number of channels produced by the convolution
        kernel_size (int or tuple): Size of the convolving kernel
        stride (int or tuple, optional): Stride of the convolution. Default: 1
        padding (int or tuple, optional): Zero-padding added to the sides of the input during their respective convolutions. Default: 0
        bias (bool, optional): If ``True``, adds a learnable bias to the output. Default: ``True``
    """

    def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, bias=False):
        super(SpatioTemporalConv, self).__init__()
        kernel_size = _triple(kernel_size)
        stride = _triple(stride)
        padding = _triple(padding)
        self.temporal_spatial_conv = nn.Conv3d(in_channels, out_channels, kernel_size, stride=stride, padding=padding, bias=bias)

    def forward(self, x):
        x = self.temporal_spatial_conv(x)
        return x


INIT_ARGS = {
    'in_channels': 3,
    'out_channels': 4,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'bias': False
}

ARGS = {
    'x': torch.randn(2, 3, 8, 16, 16)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 5, dtype family float, shape (2, 4, 8, 16, 16) — shape (2, 4, 8, 16, 16) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
WindowAttention4D Ran
ran on a synthesized input · constructed an object; no result was computed
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("b560de9c9e1dc9dc")
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 an instance of its own class, WindowAttention4D: the run constructed the object and did not call it, so no result was computed. 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)

ARGS = {
    "dim": 4,
    "num_heads": 2,
    "window_size": [2, 2],
    "qkv_bias": False,
    "attn_drop": 0.0,
    "proj_drop": 0.0,
}
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 WindowAttention4D — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a WindowAttention4D 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
compute_mask Ran
ran; our contract draft was wrong, not the code
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("961af72c78cb1191")
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 (24, 16, 16) 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)

ARGS = {
    "dims": (2, 4, 6, 8),
    "window_size": (2, 2, 2, 2),
    "shift_size": (1, 1, 1, 1),
    "device": torch.device("cpu"),
}
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 (24, 16, 16) — shape (24, 16, 16) 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_param_group_no_wd Ran
ran on a synthesized input
Wayfear/BrainNetworkTransformer/source/components/optimizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ubc-tea/com-braintf, is pointer only (licence: NONE) · get_code("b93d89b36f5d3776")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ubc-tea/com-braintf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_window_size Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("34484ba7deb03520")
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) 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 torch

torch.manual_seed(0)

# Analyze the function:
# - x_size: needs to be a sequence (list/tuple) of integers representing input dimensions
# - window_size: needs to be a sequence of integers representing window dimensions
# - shift_size: optional sequence of integers representing shift amounts, or None

# The function iterates through indices of x_size and compares with window_size
# It creates lists from window_size and optionally shift_size
# Returns tuple(s) based on whether shift_size is provided

ARGS = {
    "x_size": (8, 12, 16),  # input spatial dimensions (distinct sizes)
    "window_size": (4, 4, 4),  # window dimensions
    "shift_size": (2, 2, 2),  # shift dimensions
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
init_dataloader Ran
ran; our contract draft was wrong, not the code
Wayfear/BrainNetworkTransformer/source/dataset/dataloader.py
pointer only (licence: MIT); the stored copy get_code serves, from ubc-tea/com-braintf, is pointer only (licence: NONE) · get_code("14b27b9a81d9c709")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ubc-tea/com-braintf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

logger_factory Ran
ran; our contract draft was wrong, not the code
Wayfear/BrainNetworkTransformer/source/components/logger.py
pointer only (licence: MIT); the stored copy get_code serves, from ubc-tea/com-braintf, is pointer only (licence: NONE) · get_code("0b3026b607b39b17")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ubc-tea/com-braintf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

lr_scheduler_factory Ran
ran; our contract draft was wrong, not the code
Wayfear/BrainNetworkTransformer/source/components/lr_scheduler.py
code served (permissive licence) · get_code("6d0049b1f197cbe2")
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 1, 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.

from omegaconf import DictConfig
import torch
import torch.optim as optim
import torch.nn as nn

torch.manual_seed(0)

# Create a simple model for the optimizer
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()
optimizer = optim.SGD(model.parameters(), lr=0.1)

# Create training config
training_cfg = DictConfig({
    'total_steps': 1000,
    'steps_per_epoch': 100,
})

# Create multiple lr_config variants to test different modes
lr_configs = [
    DictConfig({
        'lr': 0.1,
        'lr_scheduler': DictConfig({
            'mode': 'step',
            'base_lr': 0.1,
            'target_lr': 0.01,
            'warm_up_from': 0.01,
            'warm_up_steps': 50,
            'milestones': [0.3, 0.6, 0.9],
            'decay_factor': 0.5,
            'poly_power': 1.0,
            'lr_decay': 0.95,
        })
    }),
    DictConfig({
        'lr': 0.1,
        'lr_scheduler': DictConfig({
            'mode': 'poly',
            'base_lr': 0.1,
            'target_lr': 0.01,
            'warm_up_from': 0.01,
            'warm_up_steps': 50,
            'milestones': [0.3, 0.6, 0.9],
            'decay_factor': 0.5,
            'poly_power': 2.0,
            'lr_decay': 0.95,
        })
    }),
    DictConfig({
        'lr': 0.1,
        'lr_scheduler': DictConfig({
            'mode': 'cos',
            'base_lr': 0.1,
            'target_lr': 0.01,
            'warm_up_from': 0.01,
            'warm_up_steps': 50,
            'milestones': [0.3, 0.6, 0.9],
            'decay_factor': 0.5,
            'poly_power': 1.0,
            'lr_decay': 0.95,
        })
    }),
    DictConfig({
        'lr': 0.1,
        'lr_scheduler': DictConfig({
            'mode': 'linear',
            'base_lr': 0.1,
            'target_lr': 0.01,
            'warm_up_from': 0.01,
            'warm_up_steps': 50,
            'milestones': [0.3, 0.6, 0.9],
            'decay_factor': 0.5,
            'poly_power': 1.0,
            'lr_decay': 0.95,
        })
    }),
    DictConfig({
        'lr': 0.1,
        'lr_scheduler': DictConfig({
            'mode': 'decay',
            'base_lr': 0.1,
            'target_lr': 0.01,
            'warm_up_from': 0.01,
            'warm_up_steps': 50,
            'milestones': [0.3, 0.6, 0.9],
            'decay_factor': 0.5,
            'poly_power': 1.0,
            'lr_decay': 0.95,
        })
    }),
]

ARGS = {
    'lr_configs': lr_configs,
    'cfg': training_cfg,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
omegaconf, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — 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
optimizer_factory Ran
ran; our contract draft was wrong, not the code
Wayfear/BrainNetworkTransformer/source/components/optimizer.py
code served (permissive licence) · get_code("be45113508613473")
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 SGD. 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
from omegaconf import DictConfig
import logging

logging.basicConfig(level=logging.INFO)

torch.manual_seed(0)

# Build a simple model with Conv2d, Linear, and BatchNorm layers
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 8, kernel_size=3)
        self.bn1 = nn.BatchNorm2d(8)
        self.linear1 = nn.Linear(8, 4)
        self.bn_linear = nn.BatchNorm1d(4)
        self.linear2 = nn.Linear(4, 2)
    
    def forward(self, x):
        return x

model = SimpleModel()

# Create optimizer config using DictConfig
optimizer_config = DictConfig({
    'name': 'SGD',
    'weight_decay': 0.01,
    'no_weight_decay': True,
    'match_rule': None,
    'except_rule': None,
    'momentum': 0.9,
    'nesterov': True
})

ARGS = {
    'model': model,
    'optimizer_config': optimizer_config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
omegaconf, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SGD — a SGD output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SGD 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
optimizers_factory Ran
ran; our contract draft was wrong, not the code
Wayfear/BrainNetworkTransformer/source/components/optimizer.py
code served (permissive licence) · get_code("fd96023efb6b2b49")
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 1, 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 torch
import torch.nn as nn
from omegaconf import DictConfig
from typing import List

torch.manual_seed(0)

# Build a small model with Conv2d, Linear, and BatchNorm layers
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 8, kernel_size=3)
        self.bn1 = nn.BatchNorm2d(8)
        self.linear1 = nn.Linear(8, 4)
        self.bn2 = nn.BatchNorm1d(4)
        self.linear2 = nn.Linear(4, 2)
    
    def forward(self, x):
        return x

model = SimpleModel()

# Create optimizer configs using DictConfig
optimizer_config_1 = DictConfig({
    'name': 'SGD',
    'weight_decay': 0.0001,
    'no_weight_decay': True,
    'match_rule': None,
    'except_rule': None,
    'momentum': 0.9,
    'nesterov': True
})

optimizer_config_2 = DictConfig({
    'name': 'Adam',
    'weight_decay': 0.0001,
    'no_weight_decay': False,
    'match_rule': None,
    'except_rule': None
})

optimizer_configs = [optimizer_config_1, optimizer_config_2]

ARGS = {
    'model': model,
    'optimizer_configs': optimizer_configs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
omegaconf, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — 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
register_vcs_handler Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/_version.py
code served (permissive licence) · get_code("c55cafde08716f01")
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.

# Global HANDLERS dict that the function uses
HANDLERS = {}

# Define arguments for register_vcs_handler
ARGS = {
    "vcs": "git",
    "method": "commit"
}
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 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
run_command Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/_version.py
code served (permissive licence) · get_code("dbb37355631dabcf")
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 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 subprocess
import sys

ARGS = {
    "commands": ["echo"],
    "args": ["hello", "world"],
    "cwd": None,
    "verbose": False,
    "hide_stderr": False,
    "env": None,
}
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 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
set_file_handler Ran
ran; our contract draft was wrong, not the code
Wayfear/BrainNetworkTransformer/source/components/logger.py
pointer only (licence: MIT); the stored copy get_code serves, from ubc-tea/com-braintf, is pointer only (licence: NONE) · get_code("b16e48b96d32841d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ubc-tea/com-braintf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

sigmoid_focal_loss Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/losses/focal_loss.py
code served (permissive licence) · get_code("0a9007867090337a")
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
import torch.nn.functional as F

torch.manual_seed(0)

# Construct input and target tensors
# input: logits, shape (batch_size, num_classes)
# target: binary labels, shape (batch_size, num_classes)
batch_size = 2
num_classes = 4

input_tensor = torch.randn(batch_size, num_classes)
target_tensor = torch.randint(0, 2, (batch_size, num_classes)).float()

# gamma: focusing parameter (default 2.0)
gamma = 2.0

# alpha: class weighting parameter - can be None, float, or tensor
# Test with a tensor of per-class weights
alpha_tensor = torch.tensor([0.25, 0.25, 0.25, 0.25])

ARGS = {
    "input": input_tensor,
    "target": target_tensor,
    "gamma": gamma,
    "alpha": alpha_tensor,
}
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
soft_dilate Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/losses/cldice.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yang-ze-kang/CORAL, is pointer only (licence: NOASSERTION) · get_code("edd33c56dd0421f0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yang-ze-kang/CORAL, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

soft_erode Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/losses/cldice.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yang-ze-kang/CORAL, is pointer only (licence: NOASSERTION) · get_code("ebb4c26db0d1a569")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yang-ze-kang/CORAL, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

soft_open Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/losses/cldice.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yang-ze-kang/CORAL, is pointer only (licence: NOASSERTION) · get_code("302a88a3b3b24caf")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yang-ze-kang/CORAL, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

softmax_focal_loss Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/losses/focal_loss.py
code served (permissive licence) · get_code("cf84889126420b19")
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)

# Construct tiny tensors
batch_size = 2
num_classes = 4

# input: unnormalized logits of shape (batch_size, num_classes)
input_tensor = torch.randn(batch_size, num_classes)

# target: ground truth labels, typically one-hot encoded or class indices
# Looking at the code, it's used as: target.shape[1] (number of classes)
# and multiplied element-wise with loss, so it should be (batch_size, num_classes)
target_tensor = torch.zeros(batch_size, num_classes)
target_tensor[0, 0] = 1.0  # one-hot encode class 0 for first sample
target_tensor[1, 2] = 1.0  # one-hot encode class 2 for second sample

# gamma: focusing parameter
gamma_val = 2.0

# alpha: can be None, a float, or a tensor
# Test with a tensor of per-class weights
alpha_tensor = torch.tensor([0.25, 0.25, 0.25, 0.25])

ARGS = {
    "input": input_tensor,
    "target": target_tensor,
    "gamma": gamma_val,
    "alpha": alpha_tensor,
}
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
spatial_gradient Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/losses/deform.py
code served (permissive licence) · get_code("d7c2de9a1750ea2b")
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, 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)

# Create a small tensor with shape BCH(WD)
# B=2 (batch), C=3 (channels), H=4 (height), W=5 (width)
x = torch.randn(2, 3, 4, 5)

# dim can be any dimension from 0 to x.ndim-1
# Let's test with dim=2 (height dimension)
dim = 2

ARGS = {
    "x": x,
    "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 4, dtype family float, shape (2, 3, 2, 3) — shape (2, 3, 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
versions_from_parentdir Ran
ran; our contract draft was wrong, not the code
Project-MONAI/MONAI/monai/_version.py
code served (permissive licence) · get_code("5032e3b81c0fffb3")
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 os
import tempfile

ARGS = {
    'parentdir_prefix': 'myproject-',
    'root': '/tmp/myproject-1.2.3/src',
    'verbose': False
}
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 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
window_reverse Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("277cca9a3532b997")
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) and returned Tensor, ndim 6, dtype family float, shape (2, 4, 6, 8, 10, 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)

# Parameters for window_reverse
b = 2  # batch size
d = 4  # depth dimension
h = 6  # height dimension
w = 8  # width dimension
t = 10  # time dimension
c = 3  # channel dimension

window_size = [2, 2, 2, 2]  # window sizes for [d, h, w, t]

# Create windows tensor
# Shape should be (B*num_windows, window_size[0], window_size[1], window_size[2], window_size[3], C)
num_windows_d = d // window_size[0]  # 2
num_windows_h = h // window_size[1]  # 3
num_windows_w = w // window_size[2]  # 4
num_windows_t = t // window_size[3]  # 5
total_windows = num_windows_d * num_windows_h * num_windows_w * num_windows_t * b

windows = torch.randn(
    b * num_windows_d * num_windows_h * num_windows_w * num_windows_t,
    window_size[0],
    window_size[1],
    window_size[2],
    window_size[3],
    c
)

dims = (b, d, h, w, t)

ARGS = {
    "windows": windows,
    "window_size": window_size,
    "dims": dims
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 6, dtype family float, shape (2, 4, 6, 8, 10, 3) — shape (2, 4, 6, 8, 10, 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
BaseModel Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("7c145bb8dd7de3e8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

BasicLayer Not yet run
unverified (harvested, no recorded run)
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("64670e8bbebf37ee")
How it ran

no invocation recorded.

BasicLayer_FullAttention Not yet run
unverified (harvested, no recorded run)
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("82954f2e594f9808")
How it ran

no invocation recorded.

BottleNeck_in Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("0f08ae993a0c2db4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

BottleNeck_out Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("fcbcaa008dd46021")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Decoder Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("c921c50ee314871e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Encoder Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("7b86cf23e6c30cc6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Encoder_Transformer_Decoder Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("453a114ddaa22943")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

R3DNet Not yet run
unverified (harvested, no recorded run)
DAVEISHAN/TCLR/tclr_pretraining/r3d_prp.py
code served (permissive licence) · get_code("8d8eac234638f60f")
How it ran

no invocation recorded.

SpatioTemporalResBlock Not yet run
unverified (harvested, no recorded run)
DAVEISHAN/TCLR/tclr_pretraining/r3d_prp.py
code served (permissive licence) · get_code("065d6753334a893e")
How it ran

no invocation recorded.

SpatioTemporalResLayer Not yet run
unverified (harvested, no recorded run)
DAVEISHAN/TCLR/tclr_pretraining/r3d_prp.py
code served (permissive licence) · get_code("dae33da258ab777e")
How it ran

no invocation recorded.

SwinTransformer4D Not yet run
unverified (harvested, no recorded run)
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("282c605afc9a9df9")
How it ran

no invocation recorded.

SwinTransformerBlock4D Not yet run
unverified (harvested, no recorded run)
transconnectome/swift/project/module/models/swin4d_transformer_ver7.py
code served (permissive licence) · get_code("32ee991678bfe389")
How it ran

no invocation recorded.

Transformer_Block Not yet run
unverified (harvested, no recorded run)
GonyRosenman/TFF/model.py
pointer only (licence: NONE) · get_code("524f03b0d35aaa91")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GonyRosenman/TFF when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

init_stratified_dataloader Not yet run
unverified (harvested, no recorded run)
Wayfear/BrainNetworkTransformer/source/dataset/dataloader.py
pointer only (licence: MIT); the stored copy get_code serves, from ubc-tea/com-braintf, is pointer only (licence: NONE) · get_code("cd853ca321de5189")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ubc-tea/com-braintf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

load_abcd_data Not yet run
unverified (harvested, no recorded run)
Wayfear/BrainNetworkTransformer/source/dataset/abcd.py
code served (permissive licence) · get_code("b660db87ab516861")
How it ran

no invocation recorded.

load_abide_data Not yet run
unverified (harvested, no recorded run)
Wayfear/BrainNetworkTransformer/source/dataset/abide.py
pointer only (licence: MIT); the stored copy get_code serves, from ubc-tea/com-braintf, is pointer only (licence: NONE) · get_code("a179c3e7ac1bdec0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ubc-tea/com-braintf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 42 samples lifted from this paper's linked repositories, 42 carry a function identity and 0 do not. Of the 42 distinct functions with an identity, 13 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 6 the same month, 7 later) and 29 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
get_window_size
brick of this paper's assembly: helper
26 other papers
12 dated earlier, 14 later · 7 only as an assembly brick
2103.14030 (2021-03)
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
2106.13230 (2021-06)
Video Swin Transformer
2111.14821 (2021-11)
End-to-End Referring Video Object Segmentation with Multimodal Transformers
and 23 more not named here
register_vcs_handler 26 other papers
22 dated earlier, 4 later
1806.00550 (2018-06)
A Swiss Army Infinitesimal Jackknife
1806.03146 (2018-06)
Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials
1806.03720 (2018-06)
Stochastic seismic waveform inversion using generative adversarial networks as a geological prior
and 23 more not named here
versions_from_parentdir 9 other papers
8 dated earlier, 1 later
1806.03146 (2018-06)
Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials
2104.08691 (2021-04)
The Power of Scale for Parameter-Efficient Prompt Tuning
2204.04516 (2022-04)
Uncertainty-Informed Deep Learning Models Enable High-Confidence Predictions for Digital Histopathology
and 6 more not named here
run_command 7 other papers
6 dated earlier, 1 later
1806.03146 (2018-06)
Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials
2104.08691 (2021-04)
The Power of Scale for Parameter-Efficient Prompt Tuning
2204.04516 (2022-04)
Uncertainty-Informed Deep Learning Models Enable High-Confidence Predictions for Digital Histopathology
and 4 more not named here
get_param_group_no_wd 1 other paper
1 the same month
2307.10181 (2023-07)
Community-Aware Transformer for Autism Prediction in fMRI Connectome
init_dataloader 1 other paper
1 the same month
2307.10181 (2023-07)
Community-Aware Transformer for Autism Prediction in fMRI Connectome
init_stratified_dataloader 1 other paper
1 the same month
2307.10181 (2023-07)
Community-Aware Transformer for Autism Prediction in fMRI Connectome
load_abide_data 1 other paper
1 the same month
2307.10181 (2023-07)
Community-Aware Transformer for Autism Prediction in fMRI Connectome
logger_factory 1 other paper
1 the same month
2307.10181 (2023-07)
Community-Aware Transformer for Autism Prediction in fMRI Connectome
set_file_handler 1 other paper
1 the same month
2307.10181 (2023-07)
Community-Aware Transformer for Autism Prediction in fMRI Connectome
soft_dilate 1 other paper
1 later
2608.30768 (2026-08)
CORAL: A Benchmark for Structure-aware and Brain-wide Neuron Reconstruction in Light Microscopy
soft_erode 1 other paper
1 later
2608.30768 (2026-08)
CORAL: A Benchmark for Structure-aware and Brain-wide Neuron Reconstruction in Light Microscopy
soft_open 1 other paper
1 later
2608.30768 (2026-08)
CORAL: A Benchmark for Structure-aware and Brain-wide Neuron Reconstruction in Light Microscopy

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

Modeling spatiotemporal brain dynamics from high-dimensional data, such as functional Magnetic Resonance Imaging (fMRI), is a formidable task in neuroscience. Existing approaches for fMRI analysis utilize hand-crafted features, but the process of feature extraction risks losing essential information in fMRI scans. To address this challenge, we present SwiFT (Swin 4D fMRI Transformer), a Swin Transformer architecture that can learn brain dynamics directly from fMRI volumes in a memory and computation-efficient manner. SwiFT achieves this by implementing a 4D window multi-head self-attention mechanism and absolute positional embeddings. We evaluate SwiFT using multiple large-scale resting-state fMRI datasets, including the Human Connectome Project (HCP), Adolescent Brain Cognitive Development (ABCD), and UK Biobank (UKB) datasets, to predict sex, age, and cognitive intelligence. Our experimental outcomes reveal that SwiFT consistently outperforms recent state-of-the-art models. Furthermore, by leveraging its end-to-end learning capability, we show that contrastive loss-based self-supervised pre-training of SwiFT can enhance performance on downstream tasks. Additionally, we employ an explainable AI method to identify the brain regions associated with sex classification. To our knowledge, SwiFT is the first Swin Transformer architecture to process dimensional spatiotemporal brain functional data in an end-to-end fashion. Our work holds substantial potential in facilitating scalable learning of functional brain imaging in neuroscience research by reducing the hurdles associated with applying Transformer models to high-dimensional fMRI.

For agents

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

get_harvested_code_for_paper("2307.05916")
get_code_for_paper("2307.05916")
have("2307.05916")

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

curl https://syntology.ai/api/ran/2307.05916.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.05916.svg)](https://syntology.ai/paper/2307.05916)

Connect an agent — have() is free.