SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.04780 · NeurIPS · 2023

IPMix: Label-Preserving Data Augmentation Method for Training Robust Classifiers

Na Zhang, Zhenglin Huang, Xianan Bao, Qingqi Zhang, Xiaomei Tu, Biao Wu, Xi Yang

arXiv · PDF · Open in the Atlas

Code that ran

15 of 17 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, 1 violated it and 0 have no contract verdict; 13 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; 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/2310.04780.json · badge: /api/ran/2310.04780.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
hzlsaber/ipmix canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
hzlsaber/IPMix marked official in Papers with Code
Papers with Code archive
13 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
get_lr Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from p3i0t/SimCLR-CIFAR10
pointer only; the stored copy get_code serves, from p3i0t/SimCLR-CIFAR10, is pointer only (licence: NONE) · get_code("190117b425ab9fe1")
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.

add Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("cbb347ae0777187b")
How it ran

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

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

# Create small test images as numpy arrays
# Images should be in range [0, 1] based on how they're used in the function
img1 = np.random.rand(2, 3, 4, 4).astype(np.float32)
img2 = np.random.rand(2, 3, 4, 4).astype(np.float32)

# Beta parameter for the beta distribution
beta = 0.5

ARGS = {
    "img1": img1,
    "img2": img2,
    "beta": beta
}
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 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
get_ab Ran
ran, honoured the contract we drafted
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("6b102ac91e94b5e7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

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

# The function get_ab takes a single parameter 'beta' which is used in np.random.beta()
# beta should be a numeric value (float or int) that represents a shape parameter

ARGS = {
    "beta": 2.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
invert Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("4cebd5ded62dc91a")
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

# Create a small PIL Image for testing
# ImageOps.invert expects a PIL Image in mode 'RGB' or 'L'
pil_img = Image.new('RGB', (4, 3), color=(100, 150, 200))

# The second parameter is unused in the function (marked with _)
unused_param = None

ARGS = {
    'pil_img': pil_img,
    '_': unused_param
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
mirror Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("68e7d0362ce44040")
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

# Create a small PIL Image for testing
pil_img = Image.new('RGB', (4, 3), color=(255, 0, 0))

# The second parameter is unused (indicated by _), so any value works
second_param = None

ARGS = {
    'pil_img': pil_img,
    '_': second_param
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
multiply Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("1c5c42ef4d6b41ed")
How it ran

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

import numpy as np

# Set random seed for reproducibility
np.random.seed(42)

# Create small numpy arrays to use as img1 and img2
img1 = np.random.rand(2, 4).astype(np.float32)
img2 = np.random.rand(2, 4).astype(np.float32)

# Create a beta parameter (float value)
beta = 2.0

ARGS = {
    'img1': img1,
    'img2': img2,
    'beta': beta
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family 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
patch_mixing Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("464e4c4f1f24b52c")
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 random
from PIL import Image
import numpy as np
from torchvision import transforms

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

# Create small PIL images for testing
def create_test_image(width, height):
    arr = np.random.randint(0, 256, (height, width, 3), dtype=np.uint8)
    return Image.fromarray(arr)

# Create a simple mixing operation function
def simple_mixing_op(tensor1, tensor2, beta):
    """Blends two tensors with beta parameter"""
    return beta * tensor1 + (1 - beta) * tensor2

# Define ARGS
ARGS = {
    'img': create_test_image(64, 64),
    'mixing_pic': create_test_image(64, 64),
    'patch_size': 32,
    'mixed_op': simple_mixing_op,
    'beta': 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
random_elems Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("198cba22a2041b0b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (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 two small image tensors with shape [C, H, W]
# Using distinct dimensions: C=3, H=4, W=5
img1 = torch.randn(3, 4, 5)
img2 = torch.randn(3, 4, 5)

# beta parameter - a scalar float
beta = 0.5

ARGS = {
    "img1": img1,
    "img2": img2,
    "beta": beta,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (3, 4, 5) — shape (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
random_pixels Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("f47705e21f50aab3")
How it ran

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

# Construct arguments for random_pixels function
# The function expects:
# - img1: tensor with shape (C, H, W)
# - img2: tensor with shape (C, H, W)
# - beta: a scalar (not used in function but required parameter)

C, H, W = 3, 4, 5
img1 = torch.rand(C, H, W)
img2 = torch.rand(C, H, W)
beta = 0.5

ARGS = {
    "img1": img1,
    "img2": img2,
    "beta": beta,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (3, 4, 5) — shape (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
rotate Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("6ff6756edee1c9da")
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 numpy as np
from PIL import Image

# Create a small PIL Image for testing
# The function expects a PIL Image object
pil_img = Image.new('RGB', (64, 64), color='red')

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

# Define arguments for the rotate function
ARGS = {
    'pil_img': pil_img,
    'level': 5
}
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 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
shear_x Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("7dbf6f8a2088e9be")
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 numpy as np
from PIL import Image
import torch

# Create a small PIL image for testing
pil_img = Image.new('RGB', (32, 32), color='red')

# Define ARGS dict with all parameters
ARGS = {
    'pil_img': pil_img,
    'level': 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
shear_y Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("ca2d43eb1a8c083f")
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 numpy as np
from PIL import Image

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

# Create a small PIL image for testing
pil_img = Image.new('RGB', (32, 32), color='red')

# Define level parameter
level = 5

# Construct the ARGS dictionary
ARGS = {
    'pil_img': pil_img,
    'level': level
}
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 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
train Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hzlsaber/ipmix/imagenet.py
code served (permissive licence) · get_code("7914dbaf0adf4d57")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned tuple, ndim 1, dtype family float; no output shape is stated. 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 and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim.lr_scheduler import StepLR

torch.manual_seed(0)

# Build a minimal model
class TinyModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 10)
    
    def forward(self, x):
        return self.fc(x)

# Build a minimal criterion
criterion = nn.CrossEntropyLoss()

# Build a minimal optimizer
model = TinyModel()
optimizer = optim.SGD(model.parameters(), lr=0.01)

# Build a minimal scheduler
scheduler = StepLR(optimizer, step_size=1, gamma=0.1)

# Build train_loader as a list of (images, targets) tuples
# For no_jsd=True case: images is a tensor, targets is a tensor
# For no_jsd=False case: images is a list of 3 tensors (clean, aug1, aug2)
train_loader_no_jsd = [
    (torch.randn(2, 4), torch.randint(0, 10, (2,))),
    (torch.randn(2, 4), torch.randint(0, 10, (2,))),
]

train_loader_with_jsd = [
    ([torch.randn(2, 4), torch.randn(2, 4), torch.randn(2, 4)], torch.randint(0, 10, (2,))),
    ([torch.randn(2, 4), torch.randn(2, 4), torch.randn(2, 4)], torch.randint(0, 10, (2,))),
]

# Build args namespace
class Args:
    def __init__(self, no_jsd=True, print_freq=1):
        self.no_jsd = no_jsd
        self.print_freq = print_freq

args_no_jsd = Args(no_jsd=True, print_freq=1)
args_with_jsd = Args(no_jsd=False, print_freq=1)

# Use the no_jsd version for the main ARGS dict
ARGS = {
    'train_loader': train_loader_no_jsd,
    'model': model,
    'criterion': criterion,
    'optimizer': optimizer,
    'scheduler': scheduler,
    'epoch': 0,
    'args': args_no_jsd,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
translate_x Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("d33ba26b62d93326")
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 numpy as np
from PIL import Image

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

# Create a simple PIL image for testing
# Create a 32x32 RGB image with some content
pil_img = Image.new('RGB', (32, 32), color=(73, 109, 137))

# Define the level parameter - should be a number that will be used in sample_level
level = 5

# Construct ARGS dictionary with all parameters
ARGS = {
    'pil_img': pil_img,
    'level': level
}
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 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
translate_y Ran
ran; our contract draft was wrong, not the code
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("0e7f36bd010453cf")
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 numpy as np
from PIL import Image

np.random.seed(0)

# Create a small PIL image for testing
pil_img = Image.new('RGB', (32, 32), color='red')

# Define arguments for translate_y function
ARGS = {
    'pil_img': pil_img,
    'level': 5
}
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 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
augment_input Not yet run
unverified (harvested, no recorded run)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("d79fa91500ab0018")
How it ran

no invocation recorded.

ipmix Not yet run
unverified (harvested, no recorded run)
hzlsaber/IPMix/cifar.py
code served (permissive licence) · get_code("4eb2047e53db0d50")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 1 later) and 11 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_lr 5 other papers
2 dated earlier, 3 later
2002.05709 (2020-02)
A Simple Framework for Contrastive Learning of Visual Representations
2112.05135 (2021-12)
PixMix: Dreamlike Pictures Comprehensively Improve Safety Measures
2312.02517 (2023-12)
Simplifying Neural Network Training Under Class Imbalance
and 2 more not named here
rotate
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1912.02781 (2019-12)
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
2108.08487 (2021-08)
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain
shear_x
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1912.02781 (2019-12)
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
2108.08487 (2021-08)
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain
shear_y
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1912.02781 (2019-12)
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
2108.08487 (2021-08)
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain
translate_x
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1912.02781 (2019-12)
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
2108.08487 (2021-08)
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain
translate_y
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1912.02781 (2019-12)
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
2108.08487 (2021-08)
Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain

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

Data augmentation has been proven effective for training high-accuracy convolutional neural network classifiers by preventing overfitting. However, building deep neural networks in real-world scenarios requires not only high accuracy on clean data but also robustness when data distributions shift. While prior methods have proposed that there is a trade-off between accuracy and robustness, we propose IPMix, a simple data augmentation approach to improve robustness without hurting clean accuracy. IPMix integrates three levels of data augmentation (image-level, patch-level, and pixel-level) into a coherent and label-preserving technique to increase the diversity of training data with limited computational overhead. To further improve the robustness, IPMix introduces structural complexity at different levels to generate more diverse images and adopts the random mixing method for multi-scale information fusion. Experiments demonstrate that IPMix outperforms state-of-the-art corruption robustness on CIFAR-C and ImageNet-C. In addition, we show that IPMix also significantly improves the other safety measures, including robustness to adversarial perturbations, calibration, prediction consistency, and anomaly detection, achieving state-of-the-art or comparable results on several benchmarks. Code is available at https://github.com/hzlsaber/IPMix.

For agents

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

get_harvested_code_for_paper("2310.04780")
get_code_for_paper("2310.04780")
have("2310.04780")

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

curl https://syntology.ai/api/ran/2310.04780.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.04780.svg)](https://syntology.ai/paper/2310.04780)

Connect an agent — have() is free.