SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1603.08511 · 2016

Colorful Image Colorization

Phillip Isola, Richard Zhang, Alexei Efros

arXiv · PDF · Open in the Atlas

Code that ran

47 of 73 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (10 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 3 violated it and 23 have no contract verdict; 19 failures of our instrument; 26 no recorded run), measured 2026-11-21; 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/1603.08511.json · badge: /api/ran/1603.08511.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
richzhang/colorization canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
Callifrey/Paddle-CIC community (classifier label)
classifier basis not recorded on the link
6 ran · 2 no recorded run
vincentbillaut/all-colors-matter community
not in the paper's declared repositories
5 ran · 4 no recorded run
demul/auto_colorization_project community
not in the paper's declared repositories
4 ran · 4 no recorded run
bluejurand/Photos-colorization community
Papers with Code archive
4 ran · 1 no recorded run
amogh7joshi/media-vision community
Papers with Code archive
4 ran · 0 no recorded run
leon-schi/ColorNet community
Papers with Code archive
4 ran · 0 no recorded run
hbenbel/Colorization community
Papers with Code archive
4 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
3 ran · 0 no recorded run
baldassarreFe/deep-koalarization community (classifier label)
classifier basis not recorded on the link
2 ran · 5 no recorded run
BerenLuthien/HyperColumns_ImageColorization community (classifier label)
classifier basis not recorded on the link
2 ran · 2 no recorded run
jkcracker/colorization community
Papers with Code archive
1 ran · 2 no recorded run
Neeleshrj/Image-Colorization-API community
Papers with Code archive
1 ran · 1 no recorded run
marumse/colorize_images community
Papers with Code archive
1 ran · 1 no recorded run
gunwoofer/projet-INF8225 community
Papers with Code archive
1 ran · 0 no recorded run
pravesh-bawangade/Image-colorization-using-Autoencoders community
not in the paper's declared repositories
1 ran · 0 no recorded run
videogorillas/colorize community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
hsalhab/Coloring-in-the-Deep community
Papers with Code archive
0 ran · 1 no recorded run
RashadGarayev/ColorizeImage community
Papers with Code archive
0 ran · 1 no recorded run
Arghyadeep/Image-Colorization-using-CNN community
Papers with Code archive
0 ran · 1 no recorded run
SharmaAjay19/VideoImageColorization community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
AdaptationNet Ran
ran on a synthesized input · constructed an object; no result was computed
demul/auto_colorization_project/pixcolor/model.py
pointer only (licence: NONE) · get_code("8e0a81c4fef9a0f0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, demul/auto_colorization_project 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.

CICECCV16Model Ran
ran on a synthesized input · constructed an object; no result was computed
amogh7joshi/media-vision/mediavision/colorize/cic/models.py
pointer only (licence: NONE) · get_code("7ef22a78a2fe85e9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, amogh7joshi/media-vision 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.

ColorEncoder Ran
ran on a synthesized input · constructed an object; no result was computed
leon-schi/ColorNet/model/model.py
pointer only (licence: NONE) · get_code("ffb6f689d84fc1b1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leon-schi/ColorNet 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.

ColorNetBuilder Ran
ran on a synthesized input · constructed an object; no result was computed
leon-schi/ColorNet/model/model.py
pointer only (licence: NONE) · get_code("e3ea91657e0463d3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leon-schi/ColorNet 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.

ColorizationNetworkv2 Ran
ran on a synthesized input · constructed an object; no result was computed
gunwoofer/projet-INF8225/colorizationNetwork.py
pointer only (licence: NONE) · get_code("c2fc10f19f130c84")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gunwoofer/projet-INF8225 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.

ColorizationUpsampling Ran
ran on a synthesized input · constructed an object; no result was computed
pravesh-bawangade/Image-colorization-using-Autoencoders/model.py
pointer only (licence: NONE) · get_code("a28f03a7f1f82de8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, pravesh-bawangade/Image-colorization-using-Autoencoders 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.

ConditioningNet Ran
ran on a synthesized input · constructed an object; no result was computed
demul/auto_colorization_project/pixcolor/model.py
pointer only (licence: NONE) · get_code("b169c1e8e95707c8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, demul/auto_colorization_project 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.

Generator256 Ran
ran on a synthesized input · constructed an object; no result was computed
hbenbel/Colorization/src/models/dcgan.py
code served (permissive licence) · get_code("2dae71a4064c1864")
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, Generator256: 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)

# Generator256 takes no parameters in __init__
ARGS = {}
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 Generator256 — 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 Generator256 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
MediaVisionModelBase Ran
ran on a synthesized input · constructed an object; no result was computed
amogh7joshi/media-vision/mediavision/colorize/cic/models.py
pointer only (licence: NONE) · get_code("8c447e0095ad2e47")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, amogh7joshi/media-vision 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.

Progbar Ran
ran on a synthesized input
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("56ddaa59e250e469")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

RGB_to_YUV Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("63cd29a5370d5d2a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

UNetBuilder Ran
ran on a synthesized input · constructed an object; no result was computed
leon-schi/ColorNet/model/model.py
pointer only (licence: NONE) · get_code("ae563a73779ff10a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leon-schi/ColorNet 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.

YUV_to_RGB Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("33037adcaa30c58f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

_conv Ran
ran; our contract draft was wrong, not the code
hbenbel/Colorization/src/models/dcgan.py
code served (permissive licence) · get_code("f26d9042fa242762")
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 Sequential. 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 = {
    "in_channels": 3,
    "out_channels": 64,
    "stride": 1,
    "kernel_size": 3,
    "padding": 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
_down_sample Ran
ran; our contract draft was wrong, not the code
hbenbel/Colorization/src/models/dcgan.py
code served (permissive licence) · get_code("f7866dcf3c966c07")
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 Sequential. 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 = {
    "in_channels": 3,
    "out_channels": 64,
    "padding": 1,
    "kernel_size": 4,
    "stride": 2,
    "negative_slope": 0.2,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
_up_sample Ran
ran; our contract draft was wrong, not the code
hbenbel/Colorization/src/models/dcgan.py
code served (permissive licence) · get_code("4c4754caddaa2f15")
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 Sequential. 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 = {
    'in_channels': 3,
    'out_channels': 16,
    'padding': 1,
    'kernel_size': 4,
    'stride': 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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
build_decoder Ran
ran; our contract draft was wrong, not the code
bluejurand/Photos-colorization/model_creation.py
pointer only (licence: NONE) · get_code("fc6d45ffb42ade2d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bluejurand/Photos-colorization 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.

build_double_convolution_block Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
amogh7joshi/media-vision/mediavision/colorize/cic/models.py
pointer only (licence: NONE) · get_code("9643de6f202f4cea")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, amogh7joshi/media-vision 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.

build_encoder Ran
ran; our contract draft was wrong, not the code
bluejurand/Photos-colorization/model_creation.py
pointer only (licence: NONE) · get_code("834a87ebdcecb372")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bluejurand/Photos-colorization 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.

build_triple_convolution_block Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
amogh7joshi/media-vision/mediavision/colorize/cic/models.py
pointer only (licence: NONE) · get_code("8efc0cab497b7f33")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, amogh7joshi/media-vision 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.

check_file Ran
ran on a synthesized input
Callifrey/Paddle-CIC/utils.py
code served (permissive licence) · get_code("33a74dbffa908d5d")
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 bool, 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 os

def check_file(file1, file2):
    filename1 = os.path.split(os.path.splitext(file1)[0])[-1]
    filename2 = os.path.split(os.path.splitext(file2)[0])[-1]
    return filename1 == filename2

ARGS = {
    "file1": "/path/to/document.txt",
    "file2": "/path/to/document.pdf"
}
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 bool, 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
check_value Ran
ran on a synthesized input
Callifrey/Paddle-CIC/resources/training_layers.py
code served (permissive licence) · get_code("416176d39fe5dd36")
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 bool, 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 numpy as np

# Test case 1: Single element array that matches the value
inds_1 = np.array([5])
val_1 = 5

# Test case 2: Single element that doesn't match
inds_2 = np.array([3])
val_2 = 7

# Test case 3: Multiple elements (should return False regardless of value)
inds_3 = np.array([1, 2, 3])
val_3 = 1

# Using the first test case for the ARGS dict
ARGS = {
    "inds": np.array([5]),
    "val": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
conv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
demul/auto_colorization_project/pixcolor/model.py
pointer only (licence: NONE) · get_code("9ed5768a61f130c3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, demul/auto_colorization_project 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.

create_image_lists Ran
ran on a synthesized input
BerenLuthien/HyperColumns_ImageColorization/read_LaMemDataset.py
code served (permissive licence) · get_code("8600a5e9c2d1e012")
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 random
import tempfile
import shutil
from tensorflow.python.platform import gfile

# Create a temporary directory structure with test images
temp_dir = tempfile.mkdtemp()
images_subdir = os.path.join(temp_dir, 'images')
os.makedirs(images_subdir, exist_ok=True)

# Create a few dummy image files
for i in range(3):
    filepath = os.path.join(images_subdir, f'test_image_{i}.jpg')
    with open(filepath, 'w') as f:
        f.write('dummy image data')

image_dir = temp_dir

ARGS = {
    'image_dir': image_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — 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
create_tl_model_embeding Ran
ran; our contract draft was wrong, not the code
bluejurand/Photos-colorization/model_creation.py
pointer only (licence: NONE) · get_code("60b0234d4cb615b6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bluejurand/Photos-colorization 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.

data_uri_to_cv2_img Ran
ran, honoured the contract we drafted
Neeleshrj/Image-Colorization-API/colorizer.py
pointer only (licence: NONE) · get_code("66e92a971dedf994")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Neeleshrj/Image-Colorization-API 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.

find_classes Ran
ran; our contract draft was wrong, not the code
Callifrey/Paddle-CIC/dataset.py
code served (permissive licence) · get_code("470a3aab197bbe05")
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 os
import tempfile
import shutil

# Create a temporary directory with subdirectories to simulate a class directory structure
temp_dir = tempfile.mkdtemp()

# Create some class subdirectories
class_names = ['class_a', 'class_b', 'class_c', 'class_d']
for class_name in class_names:
    os.makedirs(os.path.join(temp_dir, class_name), exist_ok=True)

# Define ARGS with all parameters
ARGS = {
    'dir': temp_dir,
    'classes_idx': (1, 3)
}
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 — 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
flattenData Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from philkr/magic_init
pointer only; the stored copy get_code serves, from philkr/magic_init, is pointer only (licence: NOASSERTION) · get_code("9535d5d76f73ffbf")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

forward Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from philkr/magic_init
pointer only; the stored copy get_code serves, from philkr/magic_init, is pointer only (licence: NOASSERTION) · get_code("e1c9ddd8a189ef4d")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

gatherInputData Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from philkr/magic_init
pointer only; the stored copy get_code serves, from philkr/magic_init, is pointer only (licence: NOASSERTION) · get_code("8ddb57da6b7f2943")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

gaussian Ran
ran, violated the contract we drafted
leon-schi/ColorNet/model/model.py
pointer only (licence: NONE) · get_code("ed9db0e187d4afab")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leon-schi/ColorNet 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_model_data Ran
ran on a synthesized input
BerenLuthien/HyperColumns_ImageColorization/TensorflowUtils.py
code served (permissive licence) · get_code("ecb2e0c3bd4cdd14")
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 sys
import tempfile
import scipy.io
import numpy as np

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

# Create a mock .mat file that scipy.io.loadmat can read
mock_mat_path = os.path.join(temp_dir, "mock_model.mat")
mock_data = {
    'layers': np.array([[1, 2], [3, 4]]),
    'weights': np.array([0.1, 0.2, 0.3])
}
scipy.io.savemat(mock_mat_path, mock_data)

# Extract just the filename and construct a mock URL
filename = os.path.basename(mock_mat_path)
mock_url = f"http://example.com/models/{filename}"

ARGS = {
    'dir_path': temp_dir,
    'model_url': mock_url
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, six — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_paths Ran
ran on a synthesized input
Callifrey/Paddle-CIC/utils.py
code served (permissive licence) · get_code("7995e2bd8e1c55d9")
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_root = tempfile.mkdtemp()

# Create some subdirectories with files
subdir1 = os.path.join(temp_root, "subdir1")
subdir2 = os.path.join(temp_root, "subdir2")
os.makedirs(subdir1)
os.makedirs(subdir2)

# Create nested structure in subdir1
nested1 = os.path.join(subdir1, "nested")
os.makedirs(nested1)

# Create some test files
open(os.path.join(subdir1, "file1.txt"), "w").close()
open(os.path.join(subdir1, "file2.txt"), "w").close()
open(os.path.join(nested1, "file3.txt"), "w").close()
open(os.path.join(subdir2, "file4.txt"), "w").close()

ARGS = {
    "root": temp_root
}
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
initializeWeight Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jkcracker/colorization/colorization/resources/magic_init/magic_init_mod.py
code served (permissive licence) · get_code("cabd4bcff306a546")
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 (3, 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 numpy as np

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

# Create a data matrix D with shape (N_samples, N_features)
# Need N_samples >= N_OUT for the function to work properly
N_samples = 10
N_features = 5
D = np.random.randn(N_samples, N_features)

# Number of output dimensions
N_OUT = 3

# Type of initialization
init_type = 'pca'

ARGS = {
    'D': D,
    'type': init_type,
    'N_OUT': N_OUT
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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 (3, 5) — shape (3, 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
is_image_file Ran
ran, violated the contract we drafted
Callifrey/Paddle-CIC/dataset.py
code served (permissive licence) · get_code("e12f5ee67cba16e9")
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 bool, 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

# Analyze the function is_image_file:
# - Takes 1 parameter: filename (a string)
# - Calls .lower() on it (string method)
# - Calls .endswith() on it with extensions from IMG_EXTENSIONS list
# - Returns a boolean

# The function expects a filename string that may or may not have an image extension

ARGS = {
    "filename": "example.jpg"
}
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 bool, 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
list_images Ran
ran on a synthesized input
videogorillas/colorize/colorization/runner.py
code served (permissive licence) · get_code("9e2f60b6c60fd2ae")
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 some test PNG files
temp_dir = tempfile.mkdtemp()

# Create a few test PNG files in the temporary directory
test_files = ['image1.png', 'image2.png', 'test.png']
for filename in test_files:
    filepath = os.path.join(temp_dir, filename)
    with open(filepath, 'w') as f:
        f.write("fake png content")

# Also create a non-PNG file to ensure it's filtered out
non_png_file = os.path.join(temp_dir, 'readme.txt')
with open(non_png_file, 'w') as f:
    f.write("not an image")

ARGS = {
    'in_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
load_image_jpg_to_YUV Ran
ran on a synthesized input
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("0e420dbb0c0e1e49")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

load_img Ran
ran on a synthesized input
richzhang/colorization/colorizers/util.py
code served (permissive licence) · get_code("d0c576faf88478eb")
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 (4, 3, 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 tempfile
import numpy as np
from PIL import Image
import os

# Create a temporary image file for testing
temp_dir = tempfile.mkdtemp()
img_path = os.path.join(temp_dir, "test_image.png")

# Create a simple RGB image and save it
rgb_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
img = Image.fromarray(rgb_array)
img.save(img_path)

ARGS = {
    "img_path": img_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, 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 int, shape (4, 3, 3) — shape (4, 3, 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
make_dataset Ran
ran; our contract draft was wrong, not the code
Callifrey/Paddle-CIC/dataset.py
code served (permissive licence) · get_code("afd5662a3a631d50")
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 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 os
import tempfile
import shutil

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

# Create class subdirectories with image files
class_names = ['cat', 'dog']
for class_name in class_names:
    class_dir = os.path.join(temp_dir, class_name)
    os.makedirs(class_dir, exist_ok=True)
    
    # Create some fake image files
    for i in range(2):
        img_path = os.path.join(class_dir, f'image_{i}.jpg')
        with open(img_path, 'w') as f:
            f.write('fake image data')
        
        # Also create a .png file
        png_path = os.path.join(class_dir, f'image_{i}.png')
        with open(png_path, 'w') as f:
            f.write('fake png data')
    
    # Create a non-image file that should be ignored
    non_img_path = os.path.join(class_dir, f'readme.txt')
    with open(non_img_path, 'w') as f:
        f.write('not an image')

# Create class_to_idx mapping
class_to_idx = {name: idx for idx, name in enumerate(class_names)}

ARGS = {
    'dir': temp_dir,
    'class_to_idx': class_to_idx
}
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 2, 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
make_fusion Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bluejurand/Photos-colorization/model_creation.py
pointer only (licence: NONE) · get_code("983975f7c2ea7984")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bluejurand/Photos-colorization 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.

maybe_download_inception Ran
ran on a synthesized input
baldassarreFe/deep-koalarization/src/koalarization/dataset/embedding/inception_utils.py
code served (permissive licence) · get_code("c09b7abdc79f1137")
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 str. 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 sys
import tarfile
import urllib.request
from os.path import isfile, expanduser
import tempfile
import os

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

# Create a mock checkpoint file
checkpoint_path = os.path.join(temp_dir, 'inception_resnet_v2_2016_08_30.ckpt')
with open(checkpoint_path, 'w') as f:
    f.write('mock checkpoint data')

ARGS = {
    'checkpoint_source': checkpoint_path
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
pad_image_to_size Ran
ran; our contract draft was wrong, not the code
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("c7475185fa8a1891")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

prepare_image_for_inception Ran
ran on a synthesized input
baldassarreFe/deep-koalarization/src/koalarization/dataset/embedding/inception_utils.py
code served (permissive licence) · get_code("5d365ebd19db022b")
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 SymbolicTensor. 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 tensorflow as tf
import numpy as np

# Create a small test image tensor with shape (2, 299, 299, 3) in int8 range [0, 255]
# This matches the expected input format before reshaping
input_image = tf.constant(
    np.random.randint(0, 256, size=(2, 299, 299, 3), dtype=np.uint8),
    dtype=tf.uint8
)

ARGS = {
    "input_tensor": input_image
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor 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
preprocess_img Ran
ran on a synthesized input
richzhang/colorization/colorizers/util.py
code served (permissive licence) · get_code("0b3a8d1b3bb7326d")
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 torch
import numpy as np
from skimage import color
from PIL import Image

torch.manual_seed(0)

# Create a small RGB image (numpy array with shape (height, width, 3))
# Using distinct dimensions: height=5, width=7, channels=3
img_rgb_orig = np.random.randint(0, 256, size=(5, 7, 3), dtype=np.uint8)

ARGS = {
    "img_rgb_orig": img_rgb_orig,
    "HW": (256, 256),
    "resample": 3
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, skimage, torch — 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
resize_img Ran
ran on a synthesized input
richzhang/colorization/colorizers/util.py
code served (permissive licence) · get_code("112ae2ae261d70fb")
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 (256, 256, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from PIL import Image

# Create a small test image as a numpy array
# Image dimensions: 4 rows x 3 columns x 3 channels (RGB)
img = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)

# Define arguments for resize_img function
ARGS = {
    "img": img,
    "HW": (256, 256),
    "resample": 3
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, 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 int, shape (256, 256, 3) — shape (256, 256, 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
softMaxAxis2 Ran
ran, violated the contract we drafted
marumse/colorize_images/code/classification_approach.py
pointer only (licence: NONE) · get_code("0f90e7f7b6f3f2be")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, marumse/colorize_images 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.

split_into_stacks Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
demul/auto_colorization_project/pixcolor/model.py
pointer only (licence: NONE) · get_code("eddc1dab28739f43")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, demul/auto_colorization_project 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.

ColoringModel Not yet run
unverified (harvested, no recorded run)
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("91641524db62adcf")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

Config Not yet run
unverified (harvested, no recorded run)
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("52d7e583d78fc649")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

IC_Model Not yet run
unverified (harvested, no recorded run)
hsalhab/Coloring-in-the-Deep/model.py
pointer only (licence: NONE) · get_code("5090a24732b4cd04")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, hsalhab/Coloring-in-the-Deep 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.

Image Not yet run
unverified (harvested, no recorded run)
RashadGarayev/ColorizeImage/helper.py
code served (permissive licence) · get_code("6cc55012b9cf1736")
How it ran

no invocation recorded.

PixelCNN Not yet run
unverified (harvested, no recorded run)
demul/auto_colorization_project/pixcolor/model.py
pointer only (licence: NONE) · get_code("6efa16cca8f8c1f6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, demul/auto_colorization_project 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.

UNetColoringModel Not yet run
unverified (harvested, no recorded run)
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("18e9de3390e40021")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

batch_operations Not yet run
unverified (harvested, no recorded run)
baldassarreFe/deep-koalarization/src/koalarization/dataset/tfrecords/images_queue.py
code served (permissive licence) · get_code("b09902b6a4f0d153")
How it ran

no invocation recorded.

build_complete_model Not yet run
unverified (harvested, no recorded run)
bluejurand/Photos-colorization/model_creation.py
pointer only (licence: NONE) · get_code("7faa65d5d65e6f7c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bluejurand/Photos-colorization 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.

conv_net Not yet run
unverified (harvested, no recorded run)
Arghyadeep/Image-Colorization-using-CNN/Code/colorize_cnn.py
pointer only (licence: NONE) · get_code("335ba79e7753d1ab")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Arghyadeep/Image-Colorization-using-CNN 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.

create_model Not yet run
unverified (harvested, no recorded run)
marumse/colorize_images/code/classification_approach.py
pointer only (licence: NONE) · get_code("450337b166ca479e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, marumse/colorize_images 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.

dump_YUV_image_to_jpg Not yet run
unverified (harvested, no recorded run)
vincentbillaut/all-colors-matter/models/unet.py
pointer only (licence: NONE) · get_code("bb36061ab26a8510")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vincentbillaut/all-colors-matter 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.

evaluation_pipeline Not yet run
unverified (harvested, no recorded run)
baldassarreFe/deep-koalarization/src/koalarization/training_utils.py
code served (permissive licence) · get_code("e14e85f982937fdb")
How it ran

no invocation recorded.

flatten_nd_array Not yet run
unverified (harvested, no recorded run)
Callifrey/Paddle-CIC/resources/training_layers.py
code served (permissive licence) · get_code("42539e4adb3b5d9a")
How it ran

no invocation recorded.

Showing the first 60. An agent gets every one from get_harvested_code_for_paper("1603.08511").

Also present in other papers' code

Of 73 samples lifted from this paper's linked repositories, 73 carry a function identity and 0 do not. Of the 73 distinct functions with an identity, 8 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 4 later) and 65 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
is_image_file 22 other papers
1 dated earlier, 21 later
1406.2661 (2014-06)
Generative Adversarial Networks
1805.08318 (2018-05)
Self-Attention Generative Adversarial Networks
1809.04729 (2018-09)
A Less Biased Evaluation of Out-of-distribution Sample Detectors
and 19 more not named here
MediaVisionModelBase
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2011.06294 (2020-11)
Real-Time Intermediate Flow Estimation for Video Frame Interpolation
find_classes 1 other paper
1 later
1805.08318 (2018-05)
Self-Attention Generative Adversarial Networks
flattenData
brick of this paper's assembly: helper
1 other paper
1 dated earlier
1511.06856 (2015-11)
Data-dependent Initializations of Convolutional Neural Networks
forward
brick of this paper's assembly: helper
1 other paper
1 dated earlier
1511.06856 (2015-11)
Data-dependent Initializations of Convolutional Neural Networks
gatherInputData
brick of this paper's assembly: data_transform
1 other paper
1 dated earlier
1511.06856 (2015-11)
Data-dependent Initializations of Convolutional Neural Networks
loss_with_metrics 1 other paper
1 the same month
1603.02754 (2016-03)
XGBoost: A Scalable Tree Boosting System
make_dataset 1 other paper
1 later
1805.08318 (2018-05)
Self-Attention Generative Adversarial Networks

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 2026-11-21.

Repositories linked to this paper

More are linked than shown here.

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Given a grayscale photograph as input, this paper attacks the problem of hallucinating a plausible color version of the photograph. This problem is clearly underconstrained, so previous approaches have either relied on significant user interaction or resulted in desaturated colorizations. We propose a fully automatic approach that produces vibrant and realistic colorizations. We embrace the underlying uncertainty of the problem by posing it as a classification task and use class-rebalancing at training time to increase the diversity of colors in the result. The system is implemented as a feed-forward pass in a CNN at test time and is trained on over a million color images. We evaluate our algorithm using a "colorization Turing test," asking human participants to choose between a generated and ground truth color image. Our method successfully fools humans on 32% of the trials, significantly higher than previous methods. Moreover, we show that colorization can be a powerful pretext task for self-supervised feature learning, acting as a cross-channel encoder. This approach results in state-of-the-art performance on several feature learning benchmarks.

For agents

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

get_harvested_code_for_paper("1603.08511")
get_code_for_paper("1603.08511")
have("1603.08511")

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

curl https://syntology.ai/api/ran/1603.08511.json

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

[![Syntology run record](https://syntology.ai/api/ran/1603.08511.svg)](https://syntology.ai/paper/1603.08511)

Connect an agent — have() is free.