SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1606.04797 · International Conference on 3D Vision · 2016

V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

arXiv · PDF · Open in the Atlas

Code that ran

19 of 22 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; 2 honoured the contract we drafted, 0 violated it and 16 have no contract verdict; 1 failure of our instrument; 3 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/1606.04797.json · badge: /api/ran/1606.04797.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
salem-devloper/COVID-Lung-Segment community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
YellowLight021/Vnet community (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
chakerouari/UNET_segmetation community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
salem-devloper/final community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
black0017/MedicalZooPytorch reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
alililia/VNet_Ascend community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
alexbmp/run-vnet-keras community (classifier label)
classifier basis not recorded on the link
1 ran · 2 no recorded run
mattmacy/vnet.pytorch community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
ELUCons Ran
ran; our contract draft was wrong, not the code
black0017/MedicalZooPytorch/lib/medzoo/Vnet.py
code served (permissive licence) · get_code("6fc5a148e5f07668")
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 ELU. 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 = {
    "elu": True,
    "nchan": 4
}
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 ELU — a ELU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ELU 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
accuracy_check Ran
ran on a synthesized input
chakerouari/UNET_segmetation/accuracy.py
code served (permissive licence) · get_code("7e697903c073b3a0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch
from PIL import Image

torch.manual_seed(0)

# Create two small numpy arrays with the same values for testing
mask_array = np.array([[1, 2, 3, 4],
                       [5, 6, 7, 8]], dtype=np.uint8)

prediction_array = np.array([[1, 2, 3, 4],
                             [5, 6, 7, 8]], dtype=np.uint8)

ARGS = {
    "mask": mask_array,
    "prediction": prediction_array
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
accuracy_check_for_batch Ran
ran on a synthesized input
chakerouari/UNET_segmetation/accuracy.py
code served (permissive licence) · get_code("7ab10393cbead7dc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create small batch of numpy arrays representing masks and predictions
# The function iterates through masks[index] and predictions[index]
# and calls accuracy_check on each pair
batch_size = 2

# Create small 4x3 arrays for masks and predictions
masks = [
    np.array([[1, 0, 1], [0, 1, 0], [1, 1, 0], [0, 0, 1]], dtype=np.uint8),
    np.array([[1, 1, 1], [0, 1, 0], [1, 0, 0], [0, 0, 1]], dtype=np.uint8)
]

predictions = [
    np.array([[1, 0, 1], [0, 1, 0], [1, 1, 0], [0, 0, 1]], dtype=np.uint8),
    np.array([[1, 0, 1], [0, 1, 0], [1, 0, 0], [0, 0, 1]], dtype=np.uint8)
]

ARGS = {
    'masks': masks,
    'predictions': predictions,
    'batch_size': batch_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
add_midlines Ran
ran on a synthesized input
alexbmp/run-vnet-keras/utils.py
code served (permissive licence) · get_code("f08596ef419ba0ac")
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 3, dtype family float, shape (8, 6, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small 3D numpy array to serve as input
# Using shape (8, 6, 4) to have distinct dimensions
data = np.random.rand(8, 6, 4) * 100  # Random values scaled to 0-100

ARGS = {
    "data": data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (8, 6, 4) — shape (8, 6, 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
create_annotation Ran
ran on a synthesized input
salem-devloper/COVID-Lung-Segment/convert_data.py
code served (permissive licence) · get_code("024dccab4cac2206")
How it ran

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

import os
import tempfile
import shutil

# Create a temporary directory structure with Images and Ground-truths folders
temp_dir = tempfile.mkdtemp()
images_dir = os.path.join(temp_dir, 'Images')
masks_dir = os.path.join(temp_dir, 'Ground-truths')

os.makedirs(images_dir, exist_ok=True)
os.makedirs(masks_dir, exist_ok=True)

# Create some dummy image files
image_files = ['covid_1.jpg', 'covid_2.jpg', 'normal_1.jpg', 'normal_2.jpg']
for img_file in image_files:
    with open(os.path.join(images_dir, img_file), 'w') as f:
        f.write('dummy')

# Create mask files for some images (only for covid images)
mask_files = ['mask_covid_1.jpg', 'mask_covid_2.jpg']
for mask_file in mask_files:
    with open(os.path.join(masks_dir, mask_file), 'w') as f:
        f.write('dummy')

ARGS = {
    'path': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_annotation Ran
ran on a synthesized input
salem-devloper/final/create_annotation.py
code served (permissive licence) · get_code("6d04f81ef4a995bf")
How it ran

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

import os
import tempfile
import pandas as pd

# Create a temporary directory with some dummy image files
temp_dir = tempfile.mkdtemp()
dummy_images = ['image1.jpg', 'image2.png', 'image3.jpg']
for img_name in dummy_images:
    open(os.path.join(temp_dir, img_name), 'w').close()

# Define ARGS with all required parameters
ARGS = {
    'path': temp_dir,
    'a': [0, 1, 0]  # target labels matching number of images
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
dice_coefficient Ran
ran on a synthesized input
chakerouari/UNET_segmetation/dice_loss.py
code served (permissive licence) · get_code("0d9114af7594414b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (1,) as a later run recorded it. 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, 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 batch of 2 samples, each with 4 elements
# These should be probability-like values (between 0 and 1) for dice coefficient
batch_size = 2
sample_dim = 4

input_batch = torch.rand(batch_size, sample_dim)
target_batch = torch.randint(0, 2, (batch_size, sample_dim)).float()

ARGS = {
    "input": input_batch,
    "target": target_batch
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (1,) — shape (1,) 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
dice_error Ran
ran on a synthesized input
YellowLight021/Vnet/torchloss.py
code served (permissive licence) · get_code("17de72646615f9f3")
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. 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, 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 input tensor: shape (batch_size, num_classes, ...)
# The function calls input.max(1) to get class predictions
# So we need at least 2 dimensions, with dimension 1 being the class dimension
input = torch.randn(2, 4)  # batch_size=2, num_classes=4

# Create target tensor: shape matching the batch (will be converted to FloatTensor)
# After input.max(1) and squeeze, result will have shape (2,)
# So target should also have shape (2,)
target = torch.randint(0, 2, (2,)).float()

ARGS = {
    "input": input,
    "target": target,
}
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
file_name Ran
ran on a synthesized input
alililia/VNet_Ascend/src/utils.py
code served (permissive licence) · get_code("b572a2481649012d")
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 os
import tempfile
import shutil

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

# Create some test files with 'test' and 'mhd' in the filename
test_files = [
    'test_scan_001.mhd',
    'test_scan_002.mhd',
    'other_file.txt',
    'test_data.mhd',
    'scan_003.mhd'
]

for filename in test_files:
    filepath = os.path.join(temp_dir, filename)
    with open(filepath, 'w') as f:
        f.write('dummy content')

# Define ARGS with the file_dir parameter
ARGS = {
    'file_dir': temp_dir
}
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 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
generate_masks Ran
ran on a synthesized input
salem-devloper/final/convert_data.py
code served (permissive licence) · get_code("1bf057c43ddab42d")
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 3, dtype family int, shape (8, 8, 8) 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
import numpy as np
from torch.utils.data import DataLoader

torch.manual_seed(0)

# Define a minimal segmentation network that takes images and outputs masks
class MinimalSegNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(3, 1, kernel_size=3, padding=1)
    
    def forward(self, x):
        return self.conv(x)

# Create the network
net = MinimalSegNet()

# Define a minimal dataset
class MinimalDataset:
    def __init__(self, num_samples=2, img_size=4, mask_size=4):
        self.num_samples = num_samples
        self.img_size = img_size
        self.mask_size = mask_size
    
    def __len__(self):
        return self.num_samples
    
    def __getitem__(self, idx):
        image = torch.randn(3, self.img_size, self.img_size)
        mask = torch.randint(0, 2, (1, self.mask_size, self.mask_size), dtype=torch.float32)
        return {'image': image, 'mask': mask}

# Create a simple dataloader-like structure (list of samples)
dataset = MinimalDataset(num_samples=2, img_size=4, mask_size=4)
dataloader = [dataset[i] for i in range(len(dataset))]

# Set up the device
device = 'cpu'

# Set parameters
useful_size = 8
current_size = 4

ARGS = {
    'net': net,
    'dataloader': dataloader,
    'device': device,
    'useful_size': useful_size,
    'current_size': current_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (8, 8, 8) — shape (8, 8, 8) 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_all_file_paths Ran
ran on a synthesized input
salem-devloper/COVID-Lung-Segment/croped_data.py
code served (permissive licence) · get_code("05f8948471f6b376")
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 os
import tempfile
import shutil

# Create a temporary directory structure for testing
temp_dir = tempfile.mkdtemp()

# Create some nested directories and files
os.makedirs(os.path.join(temp_dir, "subdir1"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "subdir2", "nested"), exist_ok=True)

# Create some dummy files
open(os.path.join(temp_dir, "file1.txt"), "w").close()
open(os.path.join(temp_dir, "file2.py"), "w").close()
open(os.path.join(temp_dir, "subdir1", "file3.txt"), "w").close()
open(os.path.join(temp_dir, "subdir2", "file4.dat"), "w").close()
open(os.path.join(temp_dir, "subdir2", "nested", "file5.log"), "w").close()

ARGS = {
    "directory": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — 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
get_all_file_paths Ran
ran on a synthesized input
salem-devloper/final/create_zip_file.py
code served (permissive licence) · get_code("0b888ac62a0e9b95")
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 os
import tempfile
import shutil

# Create a temporary directory structure for testing
temp_dir = tempfile.mkdtemp()

# Create some nested directories and files
os.makedirs(os.path.join(temp_dir, "subdir1"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "subdir2"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "subdir1", "nested"), exist_ok=True)

# Create some test files
with open(os.path.join(temp_dir, "file1.txt"), "w") as f:
    f.write("test")
with open(os.path.join(temp_dir, "subdir1", "file2.txt"), "w") as f:
    f.write("test")
with open(os.path.join(temp_dir, "subdir1", "nested", "file3.txt"), "w") as f:
    f.write("test")
with open(os.path.join(temp_dir, "subdir2", "file4.txt"), "w") as f:
    f.write("test")

ARGS = {
    "directory": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — 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
linear_warmup_learning_rate Ran
ran, honoured the contract we drafted
alililia/VNet_Ascend/src/utils.py
code served (permissive licence) · get_code("d6a99ec714147ac1")
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
import numpy as np

torch.manual_seed(0)

# Analyzing the function:
# - current_step: used in arithmetic (int or float)
# - warmup_steps: used in division (int or float)
# - base_lr: used in float() conversion and subtraction (int or float)
# - init_lr: used in float() conversion and addition (int or float)

ARGS = {
    "current_step": 500,
    "warmup_steps": 1000,
    "base_lr": 0.1,
    "init_lr": 0.001,
}
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 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
mask_to_image Ran
ran on a synthesized input
salem-devloper/COVID-Lung-Segment/predict.py
code served (permissive licence) · get_code("438eb3c1325aef0d")
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 int, 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 numpy as np

# The function mask_to_image expects a mask parameter
# that supports multiplication by 255 and astype conversion
# This suggests it should be a numpy array with float values in range [0, 1]

# Create a small mask array with shape (2, 4) to match the "tiny" requirement
mask = np.array([
    [0.0, 0.25, 0.5, 0.75],
    [1.0, 0.2, 0.8, 0.1]
], dtype=np.float32)

ARGS = {
    "mask": mask
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family int, 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
passthrough Ran
ran, honoured the contract we drafted
black0017/MedicalZooPytorch/lib/medzoo/Vnet.py
code served (permissive licence) · get_code("a8c5d37974c09442")
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)

# The passthrough function simply returns x unchanged
# It accepts x as a positional argument and **kwargs for any additional keyword arguments
# The function body doesn't use kwargs, so we can pass any valid value for x

# Create a small tensor for x
x = torch.randn(2, 4)

# Define ARGS dict with the required parameter
ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 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
predict Ran
ran on a synthesized input
salem-devloper/COVID-Lung-Segment/predict.py
code served (permissive licence) · get_code("9711482ce4405f41")
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 4, dtype family float, shape (2, 1, 4, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a minimal nn.Module that takes imgs and outputs masks
class SimpleSegmentationNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(3, 1, kernel_size=1)
    
    def forward(self, x):
        return self.conv(x)

# Construct arguments
imgs = torch.randn(2, 3, 4, 5)  # batch=2, channels=3, height=4, width=5
net = SimpleSegmentationNet()
device = torch.device('cpu')

ARGS = {
    'imgs': imgs,
    'net': net,
    'device': device
}
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 ndarray, ndim 4, dtype family float, shape (2, 1, 4, 5) — shape (2, 1, 4, 5) 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
rolling Ran
ran on a synthesized input
mattmacy/vnet.pytorch/plot.py
code served (permissive licence) · get_code("76cc7194ae64e0dc")
How it ran

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

import numpy as np

# Analyzing the function:
# - N: int, used as array index and divisor
# - i: array-like, indexed with slice [N-1:]
# - loss: array-like, passed to np.convolve
# - err: array-like, passed to np.convolve

# Create test data
N = 3
i = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])  # At least N elements
loss = np.array([0.5, 0.4, 0.3, 0.2, 0.1, 0.05, 0.02, 0.01, 0.005, 0.002])
err = np.array([0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.05])

ARGS = {
    'N': N,
    'i': i,
    'loss': loss,
    'err': err
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
train_test_split Ran
ran on a synthesized input
YellowLight021/Vnet/luna.py
code served (permissive licence) · get_code("c4bee74346152dc3")
How it ran

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

import numpy as np

np.random.seed(0)

# Construct sets of integers to use as test data
full = set(range(10))
positive = set(range(3, 8))
test_fraction = 0.3

ARGS = {
    "full": full,
    "positive": positive,
    "test_fraction": test_fraction,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
truncate Ran
ran on a synthesized input
YellowLight021/Vnet/luna.py
code served (permissive licence) · get_code("60f0bebd8c9d6ac0")
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 truncates values in an image (array-like)
# It modifies elements < min_bound to min_bound
# and elements > max_bound to max_bound
# Then returns the modified image

# Create a small test image as a numpy array
image = np.array([[-5.0, 0.0, 5.0, 15.0],
                   [10.0, 20.0, -10.0, 3.0]], dtype=np.float32)

# Define reasonable bounds for truncation
min_bound = 0.0
max_bound = 10.0

ARGS = {
    "image": image,
    "min_bound": min_bound,
    "max_bound": max_bound
}
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
dice_coefficient Not yet run
unverified (harvested, no recorded run)
alexbmp/run-vnet-keras/utils.py
code served (permissive licence) · get_code("4424b776451228fd")
How it ran

no invocation recorded.

dice_loss Not yet run
unverified (harvested, no recorded run)
YellowLight021/Vnet/torchloss.py
code served (permissive licence) · get_code("90323f24b4378887")
How it ran

no invocation recorded.

dice_loss Not yet run
unverified (harvested, no recorded run)
alexbmp/run-vnet-keras/utils.py
code served (permissive licence) · get_code("c42f9764fcc069fb")
How it ran

no invocation recorded.

Also present in other papers' code

Of 22 samples lifted from this paper's linked repositories, 22 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 3 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 3 later) and 18 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
linear_warmup_learning_rate 7 other papers
1 dated earlier, 1 the same month, 5 later
1506.01497 (2015-06)
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
1606.06650 (2016-06)
3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation
1904.01906 (2019-04)
What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis
and 4 more not named here
rolling 2 other papers
2 later
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1703.00443 (2017-03)
OptNet: Differentiable Optimization as a Layer in Neural Networks
dice_loss 1 other paper
1 later
1802.10508 (2018-02)
Brain Tumor Segmentation and Radiomics Survival Prediction: Contribution to the BRATS 2017 Challenge

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

Convolutional Neural Networks (CNNs) have been recently employed to solve problems from both the computer vision and medical image analysis fields. Despite their popularity, most approaches are only able to process 2D images while most medical data used in clinical practice consists of 3D volumes. In this work we propose an approach to 3D image segmentation based on a volumetric, fully convolutional, neural network. Our CNN is trained end-to-end on MRI volumes depicting prostate, and learns to predict segmentation for the whole volume at once. We introduce a novel objective function, that we optimise during training, based on Dice coefficient. In this way we can deal with situations where there is a strong imbalance between the number of foreground and background voxels. To cope with the limited number of annotated volumes available for training, we augment the data applying random non-linear transformations and histogram matching. We show in our experimental evaluation that our approach achieves good performances on challenging test data while requiring only a fraction of the processing time needed by other previous methods.

For agents

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

get_harvested_code_for_paper("1606.04797")
get_code_for_paper("1606.04797")
have("1606.04797")

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

curl https://syntology.ai/api/ran/1606.04797.json

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

[![Syntology run record](https://syntology.ai/api/ran/1606.04797.svg)](https://syntology.ai/paper/1606.04797)

Connect an agent — have() is free.