SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.06676 · 2020

Training Generative Adversarial Networks with Limited Data

arXiv · PDF · Open in the Atlas

Code that ran

21 of 29 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, 2 violated it and 15 have no contract verdict; 2 failures of our instrument; 8 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/2006.06676.json · badge: /api/ran/2006.06676.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
NVlabs/stylegan2-ada-pytorch canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
sangyun884/Face2Webtoon community (classifier label)
classifier basis not recorded on the link
9 ran · 5 no recorded run
ChristophReich1996/Multi-StyleGAN community (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
wangamelia/cmpm202p2p2 reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
beresandras/gan-flavours-keras community (classifier label)
classifier basis not recorded on the link
1 ran · 3 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_activation_statistics Ran
ran on a synthesized input
sangyun884/Face2Webtoon/gan_metrics_pytorch/fid_score.py
code served (permissive licence) · get_code("fa02bf7e7932504d")
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

torch_seed = 0
np.random.seed(torch_seed)

# Create activation statistics as a small numpy array
# Batch size: 2, Feature dimension: 4
# This represents activations from a model's layer (e.g., pool_3 from Inception)
act = np.random.randn(2, 4).astype(np.float32)

ARGS = {
    "act": act,
}
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
create_image_grid Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
wangamelia/cmpm202p2p2/generate.py
pointer only (licence: NOASSERTION) · get_code("d2790eb5fa6571da")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, wangamelia/cmpm202p2p2 (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_lenet_features Ran
ran on a synthesized input
sangyun884/Face2Webtoon/gan_metrics_pytorch/kid_score.py
code served (permissive licence) · get_code("ad1334727ae62246")
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 (200, 8) 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 numpy as np
import torch
import torch.nn as nn

torch.manual_seed(0)

class MockLeNetFeatureExtractor(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 8, kernel_size=3, padding=1)
    
    def extract_features(self, images):
        # images shape: (batch, 3, height, width)
        # Return a small feature tensor
        return self.conv1(images).mean(dim=[2, 3])  # Returns (batch, 8)

# Create a mock network
net = MockLeNetFeatureExtractor()

# Create mock image data
# Shape: (200, 3, 4, 4) - 200 images, 3 channels, 4x4 spatial dims
# Values in range [-1, 1] to trigger the normalization in the function
imgs = np.random.uniform(-1, 1, size=(200, 3, 4, 4)).astype(np.float32)

ARGS = {
    'imgs': imgs,
    'net': net,
}
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 (200, 8) — shape (200, 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
find_classes Ran
ran; our contract draft was wrong, not the code
sangyun884/Face2Webtoon/dataset.py
code served (permissive licence) · get_code("52d10a10fb72fb9f")
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 some subdirectories to simulate a class directory structure
temp_dir = tempfile.mkdtemp()
os.makedirs(os.path.join(temp_dir, "class_a"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "class_b"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "class_c"), exist_ok=True)

ARGS = {
    "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 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
generate_cut_mix_augmentation_data Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/u_net_2d_discriminator.py
code served (permissive licence) · get_code("dfbcba5442b1b006")
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 random

torch.manual_seed(0)
random.seed(0)

# Create two small distinct-sized real and fake images
# Using batch_size=2, channels=3, height=4, width=5 for distinctness
image_real = torch.randn(2, 3, 4, 5, dtype=torch.float32, device='cpu')
image_fake = torch.randn(2, 3, 4, 5, dtype=torch.float32, device='cpu')

ARGS = {
    'image_real': image_real,
    'image_fake': image_fake,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — 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
generate_cut_mix_transformation_data Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/u_net_2d_discriminator.py
code served (permissive licence) · get_code("4b65e212777706b6")
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 random

torch.manual_seed(0)
random.seed(0)

# Create small tensors with distinct dimensions
# Batch size: 2, Channels: 3, Height: 4, Width: 5
batch_size = 2
channels = 3
height = 4
width = 5

# Real image: (batch, channels, height, width)
image_real = torch.randn(batch_size, channels, height, width, device='cpu')

# Fake image: (batch, channels, height, width) - can have different batch size
image_fake = torch.randn(batch_size + 1, channels, height, width, device='cpu')

# Prediction real: (batch, 1, height, width) - pixel-wise predictions
prediction_real = torch.randn(batch_size, 1, height, width, device='cpu')

# Prediction fake: (batch, 1, height, width) - pixel-wise predictions
prediction_fake = torch.randn(batch_size + 1, 1, height, width, device='cpu')

ARGS = {
    'image_real': image_real,
    'image_fake': image_fake,
    'prediction_real': prediction_real,
    'prediction_fake': prediction_fake,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — 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
has_file_allowed_extension Ran
ran, violated the contract we drafted
sangyun884/Face2Webtoon/dataset.py
code served (permissive licence) · get_code("e800290c5ea55efd")
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.

# Setup code for has_file_allowed_extension

# The function checks if a filename ends with any of the allowed extensions
# filename: a string representing a file path
# extensions: an iterable of extension strings (like ['.jpg', '.png'])

filename = "image.jpg"
extensions = ['.jpg', '.png', '.gif', '.bmp', '.jpeg']

ARGS = {
    'filename': filename,
    'extensions': extensions
}
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
integer_translation Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/adaptive_discriminator_augmentation.py
code served (permissive licence) · get_code("3ecf4f91a6ada17a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 3, 4, 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 random

torch.manual_seed(0)
random.seed(0)

# Create a small batch of images
# Shape: (batch_size, channels, height, width)
images = torch.randn(2, 3, 4, 5)

ARGS = {
    "images": images
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 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
integer_translation Ran
ran on a synthesized input
sangyun884/Face2Webtoon/ADA.py
code served (permissive licence) · get_code("2296f6890f1691ee")
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 Image. 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
import PIL
import random
from PIL import Image
from torchvision import transforms

torch.manual_seed(0)
np.random.seed(0)
random.seed(0)

# Create a small PIL Image
pil_img = Image.new('RGB', (64, 48), color='red')

# Create arguments dictionary
ARGS = {
    'img': pil_img,
    'p': 0.7,
    'r': 0.125
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
make_dataset Ran
ran on a synthesized input
sangyun884/Face2Webtoon/dataset.py
code served (permissive licence) · get_code("273d5df61b8a0cd5")
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 with some test files
temp_dir = tempfile.mkdtemp()
os.makedirs(os.path.join(temp_dir, "subdir"), exist_ok=True)

# Create some test files with various extensions
test_files = [
    "image1.jpg",
    "image2.png",
    "image3.gif",
    "document.txt",
    "image4.JPG",
]

for fname in test_files[:3]:
    open(os.path.join(temp_dir, fname), 'w').close()

for fname in test_files[3:]:
    open(os.path.join(temp_dir, "subdir", fname), 'w').close()

# Define the arguments
ARGS = {
    "dir": temp_dir,
    "extensions": (".jpg", ".png", ".gif", ".JPG", ".PNG", ".GIF"),
    "target": 0,
}
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
normalize_0_1_batch Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/misc.py
code served (permissive licence) · get_code("d1307ae3434487a3")
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 5, dtype family float, shape (2, 2, 3, 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

torch.manual_seed(0)

# Create a small batch tensor with shape (batch_size=2, channels=3, height=4, width=5)
# The function flattens to (batch_size, -1) and then uses None indexing with 5 dimensions
# So the input must support indexing with [:, None, None, None, None]
input_tensor = torch.randn(2, 3, 4, 5)

ARGS = {
    "input": input_tensor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 5, dtype family float, shape (2, 2, 3, 4, 5) — shape (2, 2, 3, 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
normalize_m1_1_batch Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/misc.py
code served (permissive licence) · get_code("fc0c300a2ed629cf")
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 5, dtype family float, shape (2, 3, 4, 4, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a small batch tensor with shape (batch_size=2, channels=3, height=4, width=4)
# The function flattens to (batch_size, -1) and uses min/max operations
# Then reshapes back using 5 None dimensions, suggesting 5D input
input_tensor = torch.randn(2, 3, 4, 4, 2)

ARGS = {
    "input": input_tensor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 5, dtype family float, shape (2, 3, 4, 4, 2) — shape (2, 3, 4, 4, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
num_range Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from endlesssora/deceived
pointer only; the stored copy get_code serves, from endlesssora/deceived, is pointer only (licence: NOASSERTION) · get_code("6f6beb9107cb48e8")
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.

num_range Ran
ran, honoured the contract we drafted
NVlabs/stylegan2-ada-pytorch/style_mixing.py
pointer only (licence: NOASSERTION) · get_code("5c53150c611eaf49")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, NVlabs/stylegan2-ada-pytorch (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

preprocess_birds Ran
ran on a synthesized input
beresandras/gan-flavours-keras/dataset.py
code served (permissive licence) · get_code("ed196116d992670e")
How it ran

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

import tensorflow as tf
import numpy as np

# Set random seed for reproducibility
tf.random.set_seed(0)

# Create a small stand-in function that will be returned by preprocess_birds
# The function preprocess_birds returns a preprocessing function, so we need to set it up
# to be called with data dictionaries containing 'image' and 'bbox' keys

# Create sample data that matches what the preprocessing function expects
def create_sample_bird_data():
    """Create sample bird data with image and bounding box"""
    # Create a small 32x24 RGB image (height=32, width=24, channels=3)
    image = tf.constant(np.random.randint(0, 256, (32, 24, 3), dtype=np.uint8), dtype=tf.uint8)
    
    # Create a normalized bounding box in format [y_min, x_min, y_max, x_max]
    # normalized coordinates should be in [0, 1]
    bbox = tf.constant([0.1, 0.2, 0.6, 0.8], dtype=tf.float32)
    
    data = {
        'image': image,
        'bbox': bbox
    }
    return data

# Define ARGS dictionary with all parameters for preprocess_birds
ARGS = {
    'image_size': 64,
    'padding': 0.25
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
preprocessing Ran
ran on a synthesized input
sangyun884/Face2Webtoon/utils.py
code served (permissive licence) · get_code("eab7ba887c245523")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function preprocessing takes x and performs a simple normalization
# x / 127.5 - 1
# x should be a tensor or array that supports division and subtraction

# Create a small tensor with values in a typical image range (0-255)
x = torch.tensor([[100.0, 127.5, 200.0, 255.0],
                   [0.0, 50.0, 150.0, 200.0]], dtype=torch.float32)

ARGS = {
    'x': x
}
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 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
random_permutation Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/misc.py
code served (permissive licence) · get_code("1ce98174858bf663")
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 int, shape (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 numpy as np

torch.manual_seed(0)
np.random.seed(0)

ARGS = {
    "n": 5
}
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 Tensor, ndim 1, dtype family int, shape (5,) — shape (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
random_rotation90 Ran
ran on a synthesized input
sangyun884/Face2Webtoon/ADA.py
code served (permissive licence) · get_code("5377fb88325fa9b5")
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 Image. 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 PIL import Image
import random

# Create a small PIL Image for testing
# The function expects a PIL Image object
img_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
pil_img = Image.fromarray(img_array)

# Define probability parameter
p_value = 0.5

ARGS = {
    "img": pil_img,
    "p": p_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
upfirdn2d_native Ran
ran on a synthesized input
ChristophReich1996/Multi-StyleGAN/multi_stylegan/op_static/upfirdn2d.py
code served (permissive licence) · get_code("6ff3df37d07d4cf5")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 8, 6, 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.functional as F

torch.manual_seed(0)

# Construct tiny tensors with distinct dimensions
# input shape: (batch, in_h, in_w, minor)
batch_size = 2
in_h = 4
in_w = 3
minor = 5
input_tensor = torch.randn(batch_size, in_h, in_w, minor)

# kernel shape: (kernel_h, kernel_w)
kernel_h = 3
kernel_w = 3
kernel = torch.randn(kernel_h, kernel_w)

# Upsampling factors
up_x = 2
up_y = 2

# Downsampling factors
down_x = 1
down_y = 1

# Padding parameters
pad_x0 = 1
pad_x1 = 1
pad_y0 = 1
pad_y1 = 1

ARGS = {
    'input': input_tensor,
    'kernel': kernel,
    'up_x': up_x,
    'up_y': up_y,
    'down_x': down_x,
    'down_y': down_y,
    'pad_x0': pad_x0,
    'pad_x1': pad_x1,
    'pad_y0': pad_y0,
    'pad_y1': pad_y1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 8, 6, 5) — shape (2, 8, 6, 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
valmap Ran
ran, violated the contract we drafted
wangamelia/cmpm202p2p2/generate.py
pointer only (licence: NOASSERTION) · get_code("c5e25d2754ccb9a0")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, wangamelia/cmpm202p2p2 (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

x_flip Ran
ran on a synthesized input
sangyun884/Face2Webtoon/ADA.py
code served (permissive licence) · get_code("56f419ded8d6d63a")
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 Image. 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 torchvision.transforms as transforms
from PIL import Image
import numpy as np

torch.manual_seed(0)

# Create a small PIL Image to pass to x_flip
# The function expects a PIL Image as input (transforms work on PIL Images)
img_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
img = Image.fromarray(img_array.transpose(1, 2, 0))

ARGS = {
    "img": img,
    "p": 0.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
calculate_frechet_distance Not yet run
unverified (harvested, no recorded run)
sangyun884/Face2Webtoon/gan_metrics_pytorch/fid_score.py
code served (permissive licence) · get_code("0def50a351111624")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

get_activations Not yet run
unverified (harvested, no recorded run)
sangyun884/Face2Webtoon/gan_metrics_pytorch/fid_score.py
pointer only (licence: MIT); the stored copy get_code serves, from ssongwang/mix-domaincontrastivelearning, is pointer only (licence: NONE) · get_code("7b41327814ab3634")
How it ran

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

get_activations Not yet run
unverified (harvested, no recorded run)
sangyun884/Face2Webtoon/gan_metrics_pytorch/kid_score.py
code served (permissive licence) · get_code("0bdaf2a87d584365")
How it ran

no invocation recorded.

load_test_data Not yet run
unverified (harvested, no recorded run)
sangyun884/Face2Webtoon/utils.py
code served (permissive licence) · get_code("8d4ee7233484c7ee")
How it ran

no invocation recorded.

preprocess_flowers Not yet run
unverified (harvested, no recorded run)
beresandras/gan-flavours-keras/dataset.py
code served (permissive licence) · get_code("f531ed61af7fc0f2")
How it ran

no invocation recorded.

round_to_int Not yet run
unverified (harvested, no recorded run)
beresandras/gan-flavours-keras/dataset.py
code served (permissive licence) · get_code("f7831e90f18bf2e0")
How it ran

no invocation recorded.

save_images Not yet run
unverified (harvested, no recorded run)
sangyun884/Face2Webtoon/utils.py
code served (permissive licence) · get_code("5b8b7a85da8c0a00")
How it ran

no invocation recorded.

step Not yet run
unverified (harvested, no recorded run)
beresandras/gan-flavours-keras/utils.py
code served (permissive licence) · get_code("e4aa47838d197a6b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 29 samples lifted from this paper's linked repositories, 29 carry a function identity and 0 do not. Of the 28 distinct functions with an identity, 15 are also present in other papers whose code we hold (14 with at least one paper dated earlier, 1 the same month, 9 later) and 13 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
calculate_frechet_distance 74 other papers
13 dated earlier, 3 the same month, 58 later
1610.09585 (2016-10)
Conditional Image Synthesis With Auxiliary Classifier GANs
1706.08500 (2017-06)
GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
1807.11458 (2018-07)
To learn image super-resolution, use a GAN to learn how to do image degradation first
and 71 more not named here
find_classes 30 other papers
10 dated earlier, 20 later
1406.2661 (2014-06)
Generative Adversarial Networks
1612.06321 (2016-12)
Large-Scale Image Retrieval with Attentive Deep Local Features
1809.04729 (2018-09)
A Less Biased Evaluation of Out-of-distribution Sample Detectors
and 27 more not named here
has_file_allowed_extension 17 other papers
8 dated earlier, 9 later
1612.06321 (2016-12)
Large-Scale Image Retrieval with Attentive Deep Local Features
1902.02476 (2019-02)
A Simple Baseline for Bayesian Uncertainty in Deep Learning
1904.04717 (2019-04)
Label Propagation for Deep Semi-supervised Learning
and 14 more not named here
num_range 9 other papers
9 later
2106.04566 (2021-06)
Data-Efficient Instance Generation from Instance Discrimination
2106.07023 (2021-06)
Styleformer: Transformer based Generative Adversarial Networks with Style Vector
2110.08985 (2021-10)
StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis
and 6 more not named here
preprocessing 8 other papers
7 dated earlier, 1 later
1703.00848 (2017-03)
Unsupervised Image-to-Image Translation Networks
1804.04732 (2018-04)
Multimodal Unsupervised Image-to-Image Translation
1812.09912 (2018-12)
Image-to-Image Translation via Group-wise Deep Whitening-and-Coloring Transformation
and 5 more not named here
save_images 7 other papers
6 dated earlier, 1 later
1703.03208 (2017-03)
Compressed Sensing using Generative Models
1802.08406 (2018-02)
Solving Linear Inverse Problems Using GAN Priors: An Algorithm with Provable Guarantees
1811.03205 (2018-11)
Robustness of Conditional GANs to Noisy Labels
and 4 more not named here
get_activations 4 other papers
1 dated earlier, 3 later
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
2210.16559 (2022-10)
Few-shot Image Generation via Adaptation-Aware Kernel Modulation
2304.07574 (2023-04)
Exploring Incompatible Knowledge Transfer in Few-shot Image Generation
and 1 more not named here
load_test_data 4 other papers
3 dated earlier, 1 later
1703.00848 (2017-03)
Unsupervised Image-to-Image Translation Networks
1905.01270 (2019-05)
DRIT++: Diverse Image-to-Image Translation via Disentangled Representations
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
and 1 more not named here
calculate_activation_statistics 3 other papers
1 dated earlier, 2 later
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
2210.16559 (2022-10)
Few-shot Image Generation via Adaptation-Aware Kernel Modulation
2304.07574 (2023-04)
Exploring Incompatible Knowledge Transfer in Few-shot Image Generation
extract_lenet_features 1 other paper
1 dated earlier
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
get_activations 1 other paper
1 dated earlier
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
integer_translation 1 other paper
1 dated earlier
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
make_dataset 1 other paper
1 dated earlier
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
random_rotation90 1 other paper
1 dated earlier
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
x_flip 1 other paper
1 dated earlier
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation

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

Training generative adversarial networks (GAN) using too little data typically leads to discriminator overfitting, causing training to diverge. We propose an adaptive discriminator augmentation mechanism that significantly stabilizes training in limited data regimes. The approach does not require changes to loss functions or network architectures, and is applicable both when training from scratch and when fine-tuning an existing GAN on another dataset. We demonstrate, on several datasets, that good results are now possible using only a few thousand training images, often matching StyleGAN2 results with an order of magnitude fewer images. We expect this to open up new application domains for GANs. We also find that the widely used CIFAR-10 is, in fact, a limited data benchmark, and improve the record FID from 5.59 to 2.42.

For agents

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

get_harvested_code_for_paper("2006.06676")
get_code_for_paper("2006.06676")
have("2006.06676")

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

curl https://syntology.ai/api/ran/2006.06676.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.06676.svg)](https://syntology.ai/paper/2006.06676)

Connect an agent — have() is free.