SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_29065 · AAAI · 2024

Turning Waste into Wealth: Leveraging Low-Quality Samples for Enhancing Continuous Conditional Generative Adversarial Networks

Zuheng Xu, Xin Ding, Yongwei Wang

Open in the Atlas

Code that ran

18 of 18 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; 1 honoured the contract we drafted, 0 violated it and 13 have no contract verdict; 4 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/aaai_29065.json · badge: /api/ran/aaai_29065.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
UBCDingXin/Dual-NDA canonical (classifier label)
classifier basis: census
18 ran · 0 no recorded run
FunctionStatusWhere it lives
DiffAugment Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/DiffAugment_pytorch.py
pointer only (licence: MIT); the stored copy get_code serves, from SHH-Han/DNF-SR, is pointer only (licence: NOASSERTION) · get_code("1d566d068b0ed0af")
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 SHH-Han/DNF-SR, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

FID Ran
ran, honoured the contract we drafted
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/eval_metrics.py
code served (permissive licence) · get_code("b87b241eceeeab11")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned float64, 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
from scipy import linalg

np.random.seed(0)

# Create two sets of samples from multivariate Gaussians
# Xr: real data samples, shape (n_samples, n_features)
# Xg: generated data samples, shape (n_samples, n_features)

n_samples_r = 10
n_samples_g = 12
n_features = 4

# Generate random samples for real data
Xr = np.random.randn(n_samples_r, n_features).astype(np.float32)

# Generate random samples for generated data
Xg = np.random.randn(n_samples_g, n_features).astype(np.float32)

ARGS = {
    "Xr": Xr,
    "Xg": Xg,
    "eps": 1e-10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float, 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
ResNet18_class_eval Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_class_eval.py
code served (permissive licence) · get_code("7f6cb15528d5fabb")
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 ResNet_class_eval. 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 = {
    "num_classes": 5,
    "ngpu": 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 ResNet_class_eval — a ResNet_class_eval output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_class_eval 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
ResNet18_embed Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_embed.py
code served (permissive licence) · get_code("8afa0e825a06949a")
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 ResNet_embed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function ResNet18_embed takes one optional parameter: dim_embed
# We need to provide a value for it

ARGS = {
    "dim_embed": 128
}
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 ResNet_embed — a ResNet_embed output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_embed 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
ResNet18_regre_eval Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_regre_eval.py
code served (permissive licence) · get_code("db1130e2cf46cb46")
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 ResNet_regre_eval. 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)

# ResNet18_regre_eval takes one parameter: ngpu
ARGS = {
    'ngpu': 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 ResNet_regre_eval — a ResNet_regre_eval output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_regre_eval 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
ResNet34_class_eval Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_class_eval.py
code served (permissive licence) · get_code("cf33f15bc9f41ac1")
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 ResNet_class_eval. 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 = {
    "num_classes": 5,
    "ngpu": 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 ResNet_class_eval — a ResNet_class_eval output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_class_eval 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
ResNet34_embed Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_embed.py
code served (permissive licence) · get_code("9e6195961445f6df")
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 ResNet_embed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "dim_embed": 128,
}
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 ResNet_embed — a ResNet_embed output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_embed 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
ResNet34_regre_eval Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_regre_eval.py
code served (permissive licence) · get_code("7a02443c398a24c9")
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 ResNet_regre_eval. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function ResNet34_regre_eval only takes one parameter: ngpu
ARGS = {
    "ngpu": 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 ResNet_regre_eval — a ResNet_regre_eval output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_regre_eval 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
ResNet50_class_eval Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_class_eval.py
code served (permissive licence) · get_code("4403817b6bb5c37f")
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 ResNet_class_eval. 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)

# ResNet50_class_eval takes num_classes and ngpu as parameters
ARGS = {
    "num_classes": 5,
    "ngpu": 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 ResNet_class_eval — a ResNet_class_eval output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_class_eval 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
ResNet50_embed Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_embed.py
code served (permissive licence) · get_code("dd35cc77c978d33e")
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 ResNet_embed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "dim_embed": 128
}
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 ResNet_embed — a ResNet_embed output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_embed 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
ResNet50_regre_eval Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/ResNet_regre_eval.py
code served (permissive licence) · get_code("315b79a34e41c08d")
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 ResNet_regre_eval. 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)

# Build a minimal input tensor for testing
# The model expects 3-channel images
x = torch.randn(2, 3, 32, 32)  # batch_size=2, channels=3, height=32, width=32

# Define ARGS dict with all parameters for ResNet50_regre_eval
ARGS = {
    'ngpu': 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 ResNet_regre_eval — a ResNet_regre_eval output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet_regre_eval 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
normalize_images Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/eval_metrics.py
code served (permissive licence) · get_code("a2baad17144bf5f6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 3, 4, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a small batch of images with values in [0, 255] range
# Shape: (batch_size=2, channels=3, height=4, width=4)
batch_images = torch.randint(0, 256, (2, 3, 4, 4), dtype=torch.float32)

ARGS = {
    "batch_images": batch_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, 4) — shape (2, 3, 4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rand_brightness Ran
ran; our contract draft was wrong, not the code
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/DiffAugment_pytorch.py
pointer only (licence: MIT); the stored copy get_code serves, from postech-cvlab/pytorch-studiogan, is pointer only (licence: NOASSERTION) · get_code("011230b2b9b8fb6f")
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 postech-cvlab/pytorch-studiogan, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

rand_saturation Ran
ran; our contract draft was wrong, not the code
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/DiffAugment_pytorch.py
pointer only (licence: MIT); the stored copy get_code serves, from postech-cvlab/pytorch-studiogan, is pointer only (licence: NOASSERTION) · get_code("5b0d8787e670fc63")
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 postech-cvlab/pytorch-studiogan, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

snconv2d Ran
ran; our contract draft was wrong, not the code
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/sagan.py
code served (permissive licence) · get_code("572d43d0e50a2023")
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 Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.nn.utils import spectral_norm

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1,
    'groups': 1,
    'bias': True,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
snlinear Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/models/sagan.py
code served (permissive licence) · get_code("a9008b8197a342ff")
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 Linear. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.nn.utils import spectral_norm

torch.manual_seed(0)

ARGS = {
    "in_features": 4,
    "out_features": 3,
    "bias": True,
}
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 Linear — a Linear output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Linear 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
train_net_embed Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/train_net_for_label_embed.py
code served (permissive licence) · get_code("8c3b94e89bc88423")
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 EmbedNet. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define a small embedding network that returns (output, auxiliary)
class EmbedNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc1 = nn.Linear(4, 8)
        self.fc2 = nn.Linear(8, 1)
    
    def forward(self, x):
        x = self.fc1(x)
        aux = x  # auxiliary output
        x = self.fc2(x)
        return x, aux

# Create small dataloaders (list of tuples with batch images and labels)
def create_dataloader(num_batches=2, batch_size=2, input_dim=4):
    data = []
    for _ in range(num_batches):
        images = torch.randn(batch_size, input_dim)
        labels = torch.randn(batch_size)
        data.append((images, labels))
    return data

# Instantiate network and dataloaders
net = EmbedNet()
trainloader = create_dataloader(num_batches=3, batch_size=2, input_dim=4)
testloader = create_dataloader(num_batches=2, batch_size=2, input_dim=4)

# Build ARGS dict with all required parameters
ARGS = {
    'net': net,
    'net_name': 'test_embed_net',
    'trainloader': trainloader,
    'testloader': testloader,
    'epochs': 2,
    'resume_epoch': 0,
    'lr_base': 0.01,
    'lr_decay_factor': 0.1,
    'lr_decay_epochs': [1],
    'weight_decay': 0.0001,
    'path_to_ckpt': None,
}
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 EmbedNet — a EmbedNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a EmbedNet 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
train_net_y2h Ran
ran on a synthesized input
UBCDingXin/Dual-NDA/SteeringAngle/SteeringAngle_128x128/CcGAN/NDA/train_net_for_label_embed.py
code served (permissive licence) · get_code("fa83516ba2b3d962")
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 SmallY2H. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import numpy as np

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

# Create a small h2y module (hidden to y)
class SmallH2Y(nn.Module):
    def __init__(self):
        super(SmallH2Y, self).__init__()
        self.linear = nn.Linear(4, 1)
    
    def forward(self, x):
        return self.linear(x)

# Create the embed module with h2y as an attribute
class SmallEmbedModule(nn.Module):
    def __init__(self):
        super(SmallEmbedModule, self).__init__()
        self.h2y = SmallH2Y()
    
    def forward(self, x):
        return self.h2y(x)

# Create the wrapper module that has a 'module' attribute
class SmallEmbedWrapper(nn.Module):
    def __init__(self):
        super(SmallEmbedWrapper, self).__init__()
        self.module = SmallEmbedModule()
    
    def forward(self, x):
        return self.module(x)

# Create net_y2h (y to hidden)
class SmallY2H(nn.Module):
    def __init__(self):
        super(SmallY2H, self).__init__()
        self.linear = nn.Linear(1, 4)
    
    def forward(self, x):
        return self.linear(x)

# Create unique normalized labels (between 0 and 1)
unique_labels_norm = np.array([0.1, 0.3, 0.5, 0.7, 0.9], dtype=np.float32)

# Instantiate networks
net_y2h = SmallY2H()
net_embed = SmallEmbedWrapper()

ARGS = {
    'unique_labels_norm': unique_labels_norm,
    'net_y2h': net_y2h,
    'net_embed': net_embed,
    'epochs': 2,
    'lr_base': 0.01,
    'lr_decay_factor': 0.1,
    'lr_decay_epochs': [1],
    'weight_decay': 0.0001,
    'batch_size': 2
}
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 SmallY2H — a SmallY2H output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SmallY2H 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

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 3 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
rand_brightness 17 other papers
13 dated earlier, 1 later, 3 same year, order not determinable
1912.04958 (2019-12)
Analyzing and Improving the Image Quality of StyleGAN
2006.10738 (2020-06)
Differentiable Augmentation for Data-Efficient GAN Training
2007.15646 (2020-07)
Rewriting a Deep Generative Model
and 14 more not named here
rand_saturation 16 other papers
12 dated earlier, 1 later, 3 same year, order not determinable
1912.04958 (2019-12)
Analyzing and Improving the Image Quality of StyleGAN
2006.10738 (2020-06)
Differentiable Augmentation for Data-Efficient GAN Training
2007.15646 (2020-07)
Rewriting a Deep Generative Model
and 13 more not named here
DiffAugment 9 other papers
7 dated earlier, 1 later, 1 same year, order not determinable
2006.10738 (2020-06)
Differentiable Augmentation for Data-Efficient GAN Training
2104.03310 (2021-04)
Regularizing Generative Adversarial Networks under Limited Data
2111.01007 (2021-11)
Projected GANs Converge Faster
and 6 more not named here
snconv2d 4 other papers
4 dated earlier
1805.08318 (2018-05)
Self-Attention Generative Adversarial Networks
2002.06661 (2020-02)
Latent Normalizing Flows for Many-to-Many Cross-Domain Mappings
2002.11810 (2020-02)
On Leveraging Pretrained GANs for Generation with Limited Data
and 1 more not named here
FID 1 other paper
1 dated earlier
2011.07466 (2020-11)
Continuous Conditional Generative Adversarial Networks: Novel Empirical Losses and Label Input Mechanisms

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

Continuous Conditional Generative Adversarial Networks (Cc-GANs) enable generative modeling conditional on continuous scalar variables (termed regression labels). However, they can produce subpar fake images due to limited training data. Although Negative Data Augmentation (NDA) effectively enhances unconditional and class-conditional GANs by introducing anomalies into real training images, guiding the GANs away from low-quality outputs, its impact on Cc-GANs is limited, as it fails to replicate negative samples that may occur during the CcGAN sampling. We present a novel NDA approach called Dual-NDA specifically tailored for Cc-GANs to address this problem. Dual-NDA employs two types of negative samples: visually unrealistic images generated from a pre-trained CcGAN and label-inconsistent images created by manipulating real images' labels. Leveraging these negative samples, we introduce a novel discriminator objective alongside a modified CcGAN training algorithm. Empirical analysis on UTKFace and Steering Angle reveals that Dual-NDA consistently enhances the visual fidelity and label consistency of fake images generated by CcGANs, exhibiting a substantial performance gain over the vanilla NDA. Moreover, by applying Dual-NDA, CcGANs demonstrate a remarkable advancement beyond the capabilities of state-ofthe-art conditional GANs and diffusion models, establishing a new pinnacle of performance. Our codes can be found at https://github.com/UBCDingXin/Dual-NDA.

For agents

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

get_harvested_code_for_paper("aaai_29065")
get_code_for_paper("aaai_29065")
have("aaai_29065")

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

curl https://syntology.ai/api/ran/aaai_29065.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_29065.svg)](https://syntology.ai/paper/aaai_29065)

Connect an agent — have() is free.