SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1803.08494 · 2018

Group Normalization

Yuxin Wu

arXiv · PDF · Open in the Atlas

Code that ran

7 of 15 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 4 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/1803.08494.json · badge: /api/ran/1803.08494.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
ppwwyyxx/GroupNorm-reproduce canonical (classifier label)
classifier basis: census
3 ran · 1 no recorded run
labmlai/annotated_deep_learning_paper_implementations community
Papers with Code archive
1 ran · 1 no recorded run
taokong/group_normalization community
Papers with Code archive
1 ran · 0 no recorded run
dmonterom/face_recognition_TF2 community
Papers with Code archive
1 ran · 0 no recorded run
chengyangfu/pytorch-groupnormalization community
Papers with Code archive
1 ran · 0 no recorded run
jiajunhua/facebookresearch-Detectron community
Papers with Code archive
0 ran · 2 no recorded run
pren1/Group-Normalization-Example community
Papers with Code archive
0 ran · 1 no recorded run
Jarvis73/Group_Normalization community
Papers with Code archive
0 ran · 1 no recorded run
facebookresearch/Detectron community
Papers with Code archive
0 ran · 1 no recorded run
taki0112/Group_Normalization-Tensorflow community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
CIFAR10VGGModel Ran
ran on a synthesized input · constructed an object; no result was computed
labmlai/annotated_deep_learning_paper_implementations/labml_nn/normalization/group_norm/experiment.py
code served (permissive licence) · get_code("da29f85087b9fa45")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, CIFAR10VGGModel: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from typing import List

torch.manual_seed(0)

# The function signature shows __init__(self, blocks: List[List[int]])
# blocks should be a list of 5 lists, where each inner list contains channel counts
# Looking at the usage: for each block, channels are iterated and conv_blocks are created
# After each block, a MaxPool2d is applied
# Finally, fc layer maps from in_channels to 10 classes

# For CIFAR-10, typical VGG configurations use blocks like:
# [64], [128], [256, 256], [512, 512], [512, 512]
# This creates a reasonable small VGG model

blocks = [
    [64],           # Block 1
    [128],          # Block 2
    [256, 256],     # Block 3
    [512, 512],     # Block 4
    [512, 512]      # Block 5
]

ARGS = {
    "blocks": blocks
}
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 CIFAR10VGGModel — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a CIFAR10VGGModel 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
GroupBatchnorm2d Ran
ran on a synthesized input · constructed an object; no result was computed
taokong/group_normalization/group_norm.py
pointer only (licence: NONE) · get_code("b1ab93188140a94f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taokong/group_normalization when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

GroupNormalization Ran
ran on a synthesized input
dmonterom/face_recognition_TF2/resnet_groupNorm.py
pointer only (licence: NONE) · get_code("443f07c373ca43d3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dmonterom/face_recognition_TF2 when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

conv3x3 Ran
ran; our contract draft was wrong, not the code
ppwwyyxx/GroupNorm-reproduce/ImageNet-ResNet-PyTorch/resnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from lixiaotong97/DSU, 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. A run is recorded.

group_norm Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
chengyangfu/pytorch-groupnormalization/group_norm.py
pointer only (licence: NONE) · get_code("18d881658d555ef5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, chengyangfu/pytorch-groupnormalization when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

resnet18 Ran
ran; our contract draft was wrong, not the code
ppwwyyxx/GroupNorm-reproduce/ImageNet-ResNet-PyTorch/resnet.py
code served (permissive licence) · get_code("3f579c5a3b36fa45")
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. 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 = {
    'pretrained': False,
}
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 — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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 Ran
ran; our contract draft was wrong, not the code
ppwwyyxx/GroupNorm-reproduce/ImageNet-ResNet-PyTorch/resnet.py
code served (permissive licence) · get_code("e8dcd034178a8262")
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. 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 = {
    'pretrained': False
}
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 — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
GroupNorm Not yet run
unverified (harvested, no recorded run)
pren1/Group-Normalization-Example/Group_Normalization_example.py
pointer only (licence: NONE) · get_code("8def41179ee858cb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, pren1/Group-Normalization-Example 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.

GroupNorm Not yet run
unverified (harvested, no recorded run)
ppwwyyxx/GroupNorm-reproduce/Cifar10-ResNet-TensorFlow/cifar10-resnet-gn.py
code served (permissive licence) · get_code("27420f4cdbd391ae")
How it ran

no invocation recorded.

GroupNormalization Not yet run
unverified (harvested, no recorded run)
Jarvis73/Group_Normalization/GroupNormalization.py
code served (permissive licence) · get_code("bd97c13f0aaf126a")
How it ran

no invocation recorded.

Model Not yet run
unverified (harvested, no recorded run)
labmlai/annotated_deep_learning_paper_implementations/labml_nn/normalization/group_norm/experiment.py
code served (permissive licence) · get_code("e97f5db35a39c518")
How it ran

no invocation recorded.

basic_gn_stem Not yet run
unverified (harvested, no recorded run)
jiajunhua/facebookresearch-Detectron/detectron/modeling/ResNet.py
code served (permissive licence) · get_code("dd3d137d177c024b")
How it ran

no invocation recorded.

bottleneck_gn_transformation Not yet run
unverified (harvested, no recorded run)
facebookresearch/Detectron/detectron/modeling/ResNet.py
code served (permissive licence) · get_code("0bd83b7f13c89101")
How it ran

no invocation recorded.

get_group_gn Not yet run
unverified (harvested, no recorded run)
jiajunhua/facebookresearch-Detectron/detectron/modeling/ResNet.py
code served (permissive licence) · get_code("8bc93accde20e097")
How it ran

no invocation recorded.

group_norm Not yet run
unverified (harvested, no recorded run)
taki0112/Group_Normalization-Tensorflow/norm.py
code served (permissive licence) · get_code("f0a236a9978f0d8e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 3 the same month, 4 later) and 10 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
conv3x3 686 other papers
37 dated earlier, 8 the same month, 641 later · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
resnet18 89 other papers
2 dated earlier, 1 the same month, 86 later
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 86 more not named here
resnet34 53 other papers
2 dated earlier, 1 the same month, 50 later
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 50 more not named here
group_norm
brick of this paper's assembly: entry
3 other papers
1 dated earlier, 2 later
1710.10903 (2017-10)
Graph Attention Networks
2211.15281 (2022-11)
Flow: Per-instance Personalized Federated Learning
2411.00329 (2024-11)
Personalized Federated Learning via Feature Distribution Adaptation
get_group_gn
brick of this paper's assembly: helper
1 other paper
1 dated earlier · 1 only as an assembly brick
1612.03144 (2016-12)
Feature Pyramid Networks for Object Detection

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

Batch Normalization (BN) is a milestone technique in the development of deep learning, enabling various networks to train. However, normalizing along the batch dimension introduces problems -BN's error increases rapidly when the batch size becomes smaller, caused by inaccurate batch statistics estimation. This limits BN's usage for training larger models and transferring features to computer vision tasks including detection, segmentation, and video, which require small batches constrained by memory consumption. In this paper, we present Group Normalization (GN) as a simple alternative to BN. GN divides the channels into groups and computes within each group the mean and variance for normalization. GN's computation is independent of batch sizes, and its accuracy is stable in a wide range of batch sizes. On ResNet-50 trained in ImageNet, GN has 10.6% lower error than its BN counterpart when using a batch size of 2; when using typical batch sizes, GN is comparably good with BN and outperforms other normalization variants. Moreover, GN can be naturally transferred from pre-training to fine-tuning. GN can outperform its BNbased counterparts for object detection and segmentation in COCO, 1 and for video classification in Kinetics, showing that GN can effectively replace the powerful BN in a variety of tasks. GN can be easily implemented by a few lines of code in modern libraries.

For agents

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

get_harvested_code_for_paper("1803.08494")
get_code_for_paper("1803.08494")
have("1803.08494")

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

curl https://syntology.ai/api/ran/1803.08494.json

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

[![Syntology run record](https://syntology.ai/api/ran/1803.08494.svg)](https://syntology.ai/paper/1803.08494)

Connect an agent — have() is free.