Papers › ResNet strikes back: An improved training procedure in timm

ResNet strikes back: An improved training procedure in timm

1 Oct 2021NeurIPS Workshop ImageNet_PPF 2021 12arXiv:2110.00476archive 2025-07-28

Ross Wightman, Hugo Touvron, Hervé Jégou

The influential Residual Networks designed by He et al. remain the gold-standard architecture in numerous scientific publications. They typically serve as the default architecture in studies, or as baselines when new architectures are proposed. Yet there has been significant progress on best practices for training neural networks since the inception of the ResNet architecture in 2015. Novel optimization & data-augmentation have increased the effectiveness of the training recipes. In this paper, we re-evaluate the performance of the vanilla ResNet-50 when trained with a procedure that integrates such advances. We share competitive training settings and pre-trained models in the timm open-source library, with the hope that they will serve as better baselines for future work. For instance, with our more demanding training setting, a vanilla ResNet-50 reaches 80.4% top-1 accuracy at resolution 224x224 on ImageNet-val without extra data or distillation. We also report the performance achieved with popular models with our training procedure.

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

No sample from the archive's official repository (rwightman/pytorch-image-models) is linked to this paper; the 3 samples below come from 1 community repository. Syntology harvested code from rwightman/pytorch-image-models for another paper.

rwightman/pytorch-image-models: 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.

Syntology 3 code samples harvested from 1 repository linked to this paper: 0 ran · 3 unverified.

By repository: community (archive-listed): 3 samples from 1 repository, 0 ran (0 with no instrument failure; 0 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:

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

rwightman/pytorch-image-models officialmentioned in papermentioned 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
eshoyuan/taop2021-3rd-solution mentioned on GitHubpytorchSyntology: not harvested report
hunto/dist_kd mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Knowledge Distillation from A Stronger Teacher (arXiv:2205.10536; that paper's own run record: 10 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 4 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
jayroxis/ordering-in-graph-generation mentioned on GitHubpytorchCC0-1.0Syntology: 3 samples, 0 ran · 3 unverified report
open-mmlab/mmclassification mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Reversible Vision Transformers (arXiv:2302.04869; that paper's own run record: 24 ran (of which 5 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 7 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
open-mmlab/mmdetection mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to An Open and Comprehensive Pipeline for Unified Object Grounding and Detection (arXiv:2401.02361; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
samsungsailmontreal/ghn3 mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models? (arXiv:2303.04143; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
shinya7y/UniverseNet mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
sithu31296/semantic-segmentation mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers (arXiv:2105.15203; that paper's own run record: 66 ran (of which 22 constructed an object rather than computing a result; 52 with no instrument failure: 1 honoured, 2 violated, 49 with no contract checked; 14 where Syntology's instrument failed) · 20 unverified), FaPN: Feature-aligned Pyramid Network for Dense Image Prediction (arXiv:2108.07058; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified) and A ConvNet for the 2020s (arXiv:2201.03545; that paper's own run record: 57 ran (of which 39 constructed an object rather than computing a result; 49 with no instrument failure: 2 honoured, 0 violated, 47 with no contract checked; 8 where Syntology's instrument failed) · 23 unverified); nothing checks that they implement this paper's method. report
Mind23-2/MindCode-4 mindsporeSyntology: not harvested report
Westlake-AI/openmixup pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to mixup: Beyond Empirical Risk Minimization (arXiv:1710.09412; that paper's own run record: 36 ran (of which 5 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 22 where Syntology's instrument failed) · 11 unverified), Visualizing the Loss Landscape of Neural Nets (arXiv:1712.09913; that paper's own run record: 3 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified), CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features (arXiv:1905.04899; that paper's own run record: 17 ran (of which 6 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 7 where Syntology's instrument failed) · 7 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
code-implementation1/Code9 mindsporeSyntology: not harvested report
james77777778/keras-image-models pytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to GhostNetV3: Exploring the Training Strategies for Compact Models (arXiv:2404.11202; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/University mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution (arXiv:2409.12191; that paper's own run record: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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

3 samples harvested: 0 ran · 3 unverified. 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.

3unverified

Licence: 0 of the 3 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 jayroxis/ordering-in-graph-generation. “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. “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.

jayroxis/ordering-in-graph-generation community (archive-listed) 3 samples, 0 ran · 3 unverified
build_model in 469 papers model/misc.py unverified CC0-1.0 (permissive) · f8b11070de02f39b · report
build_module_registry model/misc.py unverified CC0-1.0 (permissive) · 00c7a8890d7c5be6 · report
build_partial_class model/misc.py unverified CC0-1.0 (permissive) · 35244ec766970e38 · report

Tasks

ClassificationData AugmentationDomain GeneralizationFine-Grained Image ClassificationImage ClassificationMedical Image Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Classification InDL ResNetV2_50 Average Recall 88.08% #8 of 9 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (gn) Accuracy - All Images 48.9 #9 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (gn) Accuracy - Clean Images 44.4 #9 of 90 Archive leaderboard report
Domain Generalization VizWiz-Classification ResNet-50 (gn) Accuracy - Corrupted Images 39.1 #9 of 90 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers ResNet50 (A1) Accuracy 97.9% #15 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers ResNet50 (A1) FLOPS 4.1 #15 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers ResNet50 (A1) PARAMS 24M #15 of 25 Archive leaderboard report
Fine-Grained Image Classification Stanford Cars ResNet50 (A1) Accuracy 92.7% #71 of 83 Archive leaderboard report
Fine-Grained Image Classification Stanford Cars ResNet50 (A1) FLOPS 4.1B #71 of 83 Archive leaderboard report
Fine-Grained Image Classification Stanford Cars ResNet50 (A1) PARAMS 24M #71 of 83 Archive leaderboard report
Image Classification CIFAR-10 ResNet50 (A1) Percentage correct 98.3 #47 of 265 Archive leaderboard report
Image Classification CIFAR-10 cvpr_class Percentage correct 85.28 #233 of 265 Archive leaderboard report
Image Classification CIFAR-100 ResNet50 (A1) PARAMS 25M #49 of 211 Archive leaderboard report
Image Classification CIFAR-100 ResNet50 (A1) Percentage correct 86.9 #49 of 211 Archive leaderboard report
Image Classification Flowers-102 ResNet50 (A1) Accuracy 97.9 #30 of 52 Archive leaderboard report
Image Classification Flowers-102 ResNet50 (A1) FLOPS 4.1 #30 of 52 Archive leaderboard report
Image Classification Flowers-102 ResNet50 (A1) PARAMS 25M #30 of 52 Archive leaderboard report
Image Classification ImageNet ResNet-152 (A2 + reg) Number of params 60.2M #540 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-152 (A2 + reg) Top 1 Accuracy 82.4% #540 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-152 (A2) Number of params 60.2M #610 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-152 (A2) Top 1 Accuracy 81.8% #610 of 1060 Archive leaderboard report
Image Classification ImageNet DeiT-S (T2) Number of params 22M #704 of 1060 Archive leaderboard report
Image Classification ImageNet DeiT-S (T2) Top 1 Accuracy 80.4% #704 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet50 (A1) Number of params 25M #705 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet50 (A1) Top 1 Accuracy 80.4% #705 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet50 (A3) Number of params 25M #854 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet50 (A3) Top 1 Accuracy 78.1% #854 of 1060 Archive leaderboard report
Image Classification ImageNet ReaL ResNet50 (A1) Accuracy 85.7% #39 of 57 Archive leaderboard report
Image Classification ImageNet ReaL ResNet50 (A1) Params 25M #39 of 57 Archive leaderboard report
Image Classification ImageNet V2 ResNet50 (A1) Top 1 Accuracy 68.7 #28 of 33 Archive leaderboard report
Image Classification iNaturalist 2019 ResNet50 (A2) Top-1 Accuracy 75.0 #14 of 22 Archive leaderboard report
Medical Image Classification NCT-CRC-HE-100K ResNeXt-50-32x4d Accuracy (%) 95.46 #2 of 7 Archive leaderboard report
Medical Image Classification NCT-CRC-HE-100K ResNeXt-50-32x4d F1-Score 97.46 #2 of 7 Archive leaderboard report
Medical Image Classification NCT-CRC-HE-100K ResNeXt-50-32x4d Precision 99.91 #2 of 7 Archive leaderboard report
Medical Image Classification NCT-CRC-HE-100K ResNeXt-50-32x4d Specificity 99.43 #2 of 7 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

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingKaiming InitializationMax PoolingReLUResidual BlockResidual Connection

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