Papers › Bag of Tricks for Image Classification with Convolutional Neural Networks

Bag of Tricks for Image Classification with Convolutional Neural Networks

4 Dec 2018CVPR 2019 6arXiv:1812.01187archive 2025-07-28

Tong He, Zhi Zhang, Hang Zhang, Zhongyue Zhang, Junyuan Xie, Mu Li

Much of the recent progress made in image classification research can be credited to training procedure refinements, such as changes in data augmentations and optimization methods. In the literature, however, most refinements are either briefly mentioned as implementation details or only visible in source code. In this paper, we will examine a collection of such refinements and empirically evaluate their impact on the final model accuracy through ablation study. We will show that, by combining these refinements together, we are able to improve various CNN models significantly. For example, we raise ResNet-50's top-1 validation accuracy from 75.3% to 79.29% on ImageNet. We will also demonstrate that improvement on image classification accuracy leads to better transfer learning performance in other application domains such as object detection and semantic segmentation.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

The archive's official repository (dmlc/gluon-cv) was not harvested; the 15 samples below come from 5 community repositories.

Syntology 15 code samples harvested from 5 repositories linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 2 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 15 samples from 5 repositories, 11 ran (9 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 9 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-24 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 4 of the 11 samples that ran have no recorded date.

28 repositories listed; official and paper-mentioned ones first.

dmlc/gluon-cv officialmentioned in papertfApache-2.0Syntology: not harvested report
CryptoSalamander/pytorch_paper_implementation mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Very Deep Convolutional Networks for Large-Scale Image Recognition (arXiv:1409.1556; that paper's own run record: 81 ran (of which 0 constructed an object rather than computing a result; 69 with no instrument failure: 0 honoured, 0 violated, 69 with no contract checked; 12 where Syntology's instrument failed) · 41 unverified) and Deep Residual Learning for Image Recognition (arXiv:1512.03385; that paper's own run record: 254 ran (of which 108 constructed an object rather than computing a result; 163 with no instrument failure: 3 honoured, 1 violated, 159 with no contract checked; 91 where Syntology's instrument failed) · 123 unverified); nothing checks that they implement this paper's method. report
Dmitrsl/Tools mentioned on GitHubSyntology: not harvested report
Hong-Hyun-Seung/Mask_Classification mentioned on GitHubSyntology: not harvested report
JIHOO97/Image-Classification-Competition mentioned on GitHubSyntology: not harvested report
Media-Smart/vedaseg mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to U-Net: Convolutional Networks for Biomedical Image Segmentation (arXiv:1505.04597; that paper's own run record: 530 ran (of which 247 constructed an object rather than computing a result; 378 with no instrument failure: 18 honoured, 10 violated, 350 with no contract checked; 152 where Syntology's instrument failed) · 227 unverified) and DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution,... (arXiv:1606.00915; that paper's own run record: 43 ran (of which 12 constructed an object rather than computing a result; 35 with no instrument failure: 1 honoured, 0 violated, 34 with no contract checked; 8 where Syntology's instrument failed) · 20 unverified); nothing checks that they implement this paper's method. report
PaddlePaddle/PaddleDetection mentioned on GitHubpaddleSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to PP-YOLOE: An evolved version of YOLO (arXiv:2203.16250; that paper's own run record: 22 ran (of which 0 constructed an object rather than computing a result; 19 with no instrument failure: 2 honoured, 0 violated, 17 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
PaddlePaddle/models mentioned on GitHubpaddleApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to CoKE: Contextualized Knowledge Graph Embedding (arXiv:1911.02168; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
PistonY/ResidualAttentionNetwork mentioned on GitHubtfMITSyntology: 4 samples, 3 ran (0 instrument-failed) · 1 unverified report
Tirth27/Skin-Cancer-Classification-using-Deep-Learning mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (arXiv:1905.11946; that paper's own run record: 198 ran (of which 73 constructed an object rather than computing a result; 157 with no instrument failure: 26 honoured, 2 violated, 129 with no contract checked; 41 where Syntology's instrument failed) · 104 unverified); nothing checks that they implement this paper's method. report
TotalVariation/Flattenet mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
cinastanbean/Pytorch-Multi-Task-Multi-class-Classification mentioned on GitHubpytorchSyntology: 3 samples, 1 ran (1 instrument-failed) · 2 unverified report
elasticdeeplearning/edl mentioned on GitHubpaddleApache-2.0Syntology: not harvested report
jameswang287/Car-Detection mentioned on GitHubpytorchSyntology: not harvested report
paddlepaddle/edl mentioned on GitHubpaddleApache-2.0Syntology: not harvested report
pshashk/seesaw-facenet mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SeesawFaceNets: sparse and robust face verification model for mobile platform (arXiv:1908.09124; that paper's own run record: 15 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
qingyuanchen1997/Bag-of-Tricks mentioned on GitHubpytorchMITSyntology: not harvested report
rwightman/pytorch-image-models mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights (arXiv:2006.08217; that paper's own run record: 4 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
seermer/TensorFlow2-EfficientNetV2 mentioned on GitHubtfSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
seominseok0429/pytorch-warmup-cosine-lr mentioned on GitHubpytorchSyntology: not harvested report
sherdencooper/tricks-in-deeplearning mentioned on GitHubtfSyntology: not harvested report
smilingwolf/sw-cv-modelzoo mentioned on GitHubtfSyntology: not harvested report
tyohei/chainerkfac mentioned on GitHubSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
yangyucheng000/resnet_Ascend mentioned on GitHubmindsporeApache-2.0Syntology: 5 samples, 4 ran (0 instrument-failed) · 1 unverified report
PaddlePaddle/PaddleClas paddleApache-2.0Syntology: not harvested report
PaddlePaddle/PaddleOCR paddleApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to... (arXiv:2508.19944; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 3 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
open-mmlab/mmpose pytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RTMO: Towards High-Performance One-Stage Real-Time Multi-Person Pose Estimation (arXiv:2312.07526; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/resnet_GPU mindsporeApache-2.0Syntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

15 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 2 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
2ran · our draft was wrong
7ran
4unverified

Licence: 5 of the 15 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 2 samples whose link to this paper records no repository are listed under one of this paper's own repositories, because identical code was harvested from it for another paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

PistonY/ResidualAttentionNetwork community (archive-listed) 4 samples, 3 ran (0 instrument-failed) · 1 unverified
cutout in 17 papers lib/piston_util.py ran MIT (permissive) · de47a57778dd152f · report fixture: signature p:float/0/float, s_l:float/0/float, s_h:float/0/float, r_1:float/0/float, r_2:float/0/float, v_l:int/0/int, v_h:int/0/int, pixel_level:bool/0/bool · returned function · level recorded 2026-09-26
format_time in 140 papers lib/piston_util.py ran MIT (permissive) · e83a42da4ac207cc · report fixture: signature prev_time:timedelta/0/?, cur_time:timedelta/0/? · returned str · level recorded 2026-09-26
get_ems kaggle/utils.py ran MIT (permissive) · 7d86840aa94ec126 · report fixture: signature net:list/1/callable, trans:Tensor/2/float, rate:list/1/float · returned Tensor · level recorded 2026-09-26
rescale in 30 papers lib/piston_util.py unverified MIT (permissive) · 237abd0c98440417 · report
cinastanbean/Pytorch-Multi-Task-Multi-class-Classification community (archive-listed) 3 samples, 1 ran (1 instrument-failed) · 2 unverified
get_dict_key src/main_mtmc_resnet.py ran · our draft was wrong no licence file found · pointer only · b8d5814ad4631b08 · report fixture: signature dict:dict/0/?, value:float/0/float · returned str · level recorded, date not recorded
cout_confmatrix_tofile src/main_mtmc_resnet.py unverified no licence file found · pointer only · 38093779943ae353 · report
slmc_validate src/main_mtmc_resnet.py unverified no licence file found · pointer only · bb26e0be1ece15c0 · report
seermer/TensorFlow2-EfficientNetV2 community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
round_filters in 47 papers efficientnetV2.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · honoured contract fingerprinted licence of this copy not recorded · e60387e9529f63ff · report fixture: signature filters:int/0/int, multiplier:float/0/float · returned int · level recorded, date not recorded
round_repeats in 39 papers efficientnetV2.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · honoured contract fingerprinted licence of this copy not recorded · 1a9d7e65e0a9fbc9 · report fixture: signature repeats:int/0/int, multiplier:float/0/float · returned int · level recorded, date not recorded
tyohei/chainerkfac community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
extract in 149 papers
census: 9 implementations from 12 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
chainerkfac/optimizers/kfac.py ran · our draft was wrong MIT (permissive) · b4e26744ad9e217a · report fixture: signature blocks:list/1/list, indices:list/1/list, extractor:callable/0/? · returned list · level recorded, date not recorded
yangyucheng000/resnet_Ascend community (archive-listed) 5 samples, 4 ran (0 instrument-failed) · 1 unverified
calculate_gain in 14 papers src/resnet.py ran Apache-2.0 (permissive) · 82c48515ddda4614 · report fixture: signature nonlinearity:str/0/str, param:NoneType/0/? · returned float · level recorded 2026-09-24
get_liner_lr gpu_resnet_benchmark.py ran Apache-2.0 (permissive) · 11d835e45631ea6b · report fixture: signature lr_init:float/0/float, lr_end:float/0/float, lr_max:float/0/float, warmup_epochs:int/0/int, total_epochs:int/0/int, steps_per_epoch:int/0/int · returned ndarray · level recorded 2026-09-26
get_top5_acc postprocess.py ran Apache-2.0 (permissive) · 52a9653519e980d2 · report fixture: signature top5_arg:list/1/list, gt_class:list/1/int · returned int · level recorded 2026-09-26
kaiming_normal src/resnet.py ran Apache-2.0 (permissive) · f0a6216664197738 · report fixture: signature inputs_shape:tuple/1/int, a:float/0/float, mode:str/0/str, nonlinearity:str/0/str · returned ndarray · level recorded 2026-09-24
show_predict_info infer.py unverified Apache-2.0 (permissive) · c5cd33786bbc32f6 · report

Tasks

Domain GeneralizationGeneral ClassificationImage ClassificationObject DetectionSemantic SegmentationTransfer Learningimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Domain Generalization VizWiz-Classification ResNet-26-D Accuracy - All Images 39.7 #38 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-26-D Accuracy - Clean Images 43.5 #38 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-26-D Accuracy - Corrupted Images 35.8 #38 of 90 Archive leaderboard report
Image Classification ImageNet ResNet-50-D Number of params 25M #883 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-50-D Top 1 Accuracy 77.16% #883 of 1060 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: ResNet-D

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionCosine AnnealingGlobal Average PoolingLabel SmoothingMax PoolingMixupNesterov Accelerated GradientRandom Horizontal FlipRandom Resized CropReLUResNet-DResidual BlockResidual ConnectionWeight DecayXavier Initialization

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections