當前位置：首頁 > 编程资源 > 编程问答 >内容正文

编程问答

A ConvNet for the 2020s

發布時間：2023/12/20 编程问答 31 豆豆

生活随笔收集整理的這篇文章主要介紹了 A ConvNet for the 2020s 小編覺得挺不錯的,現在分享給大家,幫大家做個參考.

作者：Zhuang Liu^1,2* Hanzi Mao¹ Chao-Yuan Wu¹ Christoph Feichtenhofer¹ Trevor Darrell² Saining Xie^1?
機構：¹Facebook AI Research (FAIR) ²UC Berkeley

*Work done during an internship at Facebook AI Research. —— 在Facebook人工智能研究部實習期間完成的工作。
?Corresponding author.

Abstract

The “Roaring 20s” of visual recognition began with the introduction of Vision Transformers (ViTs), which quickly superseded ConvNets as the state-of-the-art image classification model. A vanilla ViT, on the other hand, faces difficulties when applied to general computer vision tasks such as object detection and semantic segmentation. It is the hierarchical Transformers (e.g., Swin Transformers) that reintroduced several ConvNet priors, making Transformers practically viable as a generic vision backbone and demonstrating remarkable performance on a wide variety of vision tasks. However, the effectiveness of such hybrid approaches is still largely credited to the intrinsic superiority of Transformers, rather than the inherent inductive biases of convolutions. In this work, we reexamine the design spaces and test the limits of what a pure ConvNet can achieve. We gradually “modernize” a standard ResNet toward the design of a vision Transformer, and discover several key components that contribute to the performance difference along the way. The outcome of this exploration is a family of pure ConvNet models dubbed ConvNeXt. Constructed entirely from standard ConvNet modules, ConvNeXts compete favorably with Transformers in terms of accuracy and scalability, achieving 87.8% ImageNet top-1 accuracy and outperforming Swin Transformers on COCO detection and ADE20K segmentation, while maintaining the simplicity and efficiency of standard ConvNets.

視覺識別的 "咆哮20年代 "始于視覺Transformer（ViTs）的引入，它迅速取代了ConvNets成為最先進的圖像分類模型。另一方面，一個虛無的ViT在應用于一般的計算機視覺任務時面臨著困難，如目標檢測和語義分割。正是分層Transformer（如Swin Transformers）重新引入了幾個ConvNet先驗 (priors)，使得Transformer作為通用視覺骨干實際上是可行的，并在各種視覺任務中表現出顯著的性能。然而，這種混合方法的有效性仍然主要歸功于Transformers的內在優勢，而不是Convolutions的內在歸納偏置（the inherent inductive biases of convolutions）。在這項工作中，我們重新審視了設計空間（design spaces），并測試了純ConvNet所能實現的極限。我們逐步將一個標準的ResNet “現代化（modernize）”，使之成為一個視覺Transformer的設計，并在這一過程中發現了幾個促成性能差異的關鍵組件。這一探索的結果是一個被稱為ConvNeXt的純ConvNet模型系列。ConvNeXt完全由標準的ConvNet模塊構成，在準確性和可擴展性方面與Transformer競爭，在COCO檢測和ADE20K分割方面達到了87.8%的ImageNet top-1準確性并超過了Swin Transformers，同時保持了標準ConvNets的簡單性和效率。

這里面的ConvNets指的是基于CNN的網絡。

1. Introduction

Looking back at the 2010s, the decade was marked by the monumental progress and impact of deep learning. The primary driver was the renaissance of neural networks, particularly convolutional neural networks (ConvNets). Through the decade, the field of visual recognition successfully shifted from engineering features to designing (ConvNet) architectures. Although the invention of back-propagationtrained ConvNets dates all the way back to the 1980s [42], it was not until late 2012 that we saw its true potential for visual feature learning. The introduction of AlexNet [40] precipitated the “ImageNet moment” [59], ushering in a new era of computer vision. The field has since evolved at a rapid speed. Representative ConvNets like VGGNet [64], Inceptions [68], ResNe(X)t [28, 87], DenseNet [36], MobileNet [34], EfficientNet [71] and RegNet [54] focused on different aspects of accuracy, efficiency and scalability, and popularized many useful design principles.

回顧2010年代，這十年的特點是深度學習的巨大進步和影響。主要驅動力是神經網絡的復興，特別是卷積神經網絡（ConvNets）。在這十年中，視覺識別領域成功地從工程特征轉向設計（ConvNet）架構。雖然反向傳播訓練的ConvNets的發明可以追溯到20世紀80年代，但直到2012年底，我們才看到它在視覺特征學習方面的真正潛力。AlexNet的引入催生了 “ImageNet時刻”，開創了計算機視覺的新時代。此后，該領域以極快的速度發展起來。代表性的ConvNets如

VGGNet
Inceptions
ResNe(X)t
DenseNet
MobileNet
EfficientNet
RegNet
…

專注于準確性、效率和可擴展性的不同方面，并推廣了許多有用的設計原則。

The full dominance of ConvNets in computer vision was not a coincidence: in many application scenarios, a “sliding window” strategy is intrinsic to visual processing, particularly when working with high-resolution images. ConvNets have several built-in inductive biases that make them wellsuited to a wide variety of computer vision applications. The most important one is translation equivariance, which is a desirable property for tasks like objection detection. ConvNets are also inherently efficient due to the fact that when used in a sliding-window manner, the computations are shared [62]. For many decades, this has been the default use of ConvNets, generally on limited object categories such as digits [43], faces [58, 76] and pedestrians [19, 63]. Entering the 2010s, the region-based detectors [23, 24, 27, 57] further elevated ConvNets to the position of being the fundamental building block in a visual recognition system.

ConvNets在計算機視覺中的全面主導地位并不是一個巧合：在許多應用場景中，"滑動窗口（sliding window）"策略是視覺處理的內在因素，特別是在處理高分辨率圖像時。ConvNets有幾個內置的歸納偏置，使它們非常適合于各種計算機視覺應用。最重要的是平移等變性 (translation equivariant)，這是目標檢測等任務的一個理想屬性。ConvNets本身也是高效的，因為當以滑動窗口的方式使用時，計算是共享的（也就是常說的卷積第二個特征——權值共享）。幾十年來，這一直是ConvNets的默認用法，一般用于有限的對象類別，如數字、人臉和行人。進入2010年代，基于區域的檢測器（region-based detectors）進一步提升了ConvNets的地位，成為視覺識別系統的基本構件。

translation equivariant: 卷積操作具有平移等變性(translation equivariant)，這意味著它保存了轉換，而CNN則允許平移不變性（translation invariance）這是通過適當的(即與空間特征相關的)降維來實現的。

Around the same time, the odyssey of neural network design for natural language processing (NLP) took a very different path, as the Transformers replaced recurrent neural networks to become the dominant backbone architecture. Despite the disparity in the task of interest between language and vision domains, the two streams surprisingly converged in the year 2020, as the introduction of Vision Transformers (ViT) completely altered the landscape of network architecture design. Except for the initial “patchify” layer, which splits an image into a sequence of patches, ViT introduces no image-specific inductive bias and makes minimal changes to the original NLP Transformers. One primary focus of ViT is on the scaling behavior: with the help of larger model and dataset sizes, Transformers can outperform standard ResNets by a significant margin. Those results on image classification tasks are inspiring, but computer vision is not limited to image classification. As discussed previously, solutions to numerous computer vision tasks in the past decade depended significantly on a sliding-window, fully convolutional paradigm. Without the ConvNet inductive biases, a vanilla ViT model faces many challenges in being adopted as a generic vision backbone. The biggest challenge is ViT’s global attention design, which has a quadratic complexity with respect to the input size. This might be acceptable for ImageNet classification, but quickly becomes intractable with higher-resolution inputs.

大約在同一時間，用于自然語言處理（NLP）的神經網絡設計的漫長而充滿風險地走了一條非常不同的道路，因為Transformer取代了遞歸神經網絡（RNN），成為了主流的骨干架構。盡管語言和視覺領域的關注點任務不盡相同，但這兩股潮流在2020年出人意料地融合在一起，因為Vision Transformers（ViT）的引入完全改變了網絡架構設計的格局。除了最初的 "補丁化"層 —— patchify（將圖像分割成一連串的patches），ViT沒有引入圖像特定的歸納偏置，對原始的NLP變形器的改動也很小。ViT的一個主要關注點是擴展行為：在更大的模型和數據集規模的幫助下，Transformers可以在很大程度上超過標準ResNets的表現。這些關于圖像分類任務的結果是鼓舞人心的，但計算機視覺并不限于圖像分類。如前所述，在過去十年中，許多計算機視覺任務的解決方案在很大程度上依賴于滑動窗口、全卷積范式(fully convolutional paradigm)。如果沒有ConvNet的歸納偏置，視覺的ViT模型在作為通用視覺骨干時面臨許多挑戰。最大的挑戰是ViT的全局注意力設計，它的復雜度與輸入大小呈二次方。這對于ImageNet分類來說可能是可以接受的，但對于更高分辨率的輸入來說很快就變得難以解決了。

Hierarchical Transformers employ a hybrid approach to bridge this gap. For example, the “sliding window” strategy (e.g. attention within local windows) was reintroduced to Transformers, allowing them to behave more similarly to ConvNets. Swin Transformer [45] is a milestone work in this direction, demonstrating for the first time that Transformers can be adopted as a generic vision backbone and achieve state-of-the-art performance across a range of computer vision tasks beyond image classification. Swin Transformer’s success and rapid adoption also revealed one thing: the essence of convolution is not becoming irrelevant; rather, it remains much desired and has never faded.

分層Transformer采用了一種混合方法來彌補這一差距。例如，"滑動窗口 "策略（如在局部窗口內的注意）被重新引入Transformers，使其行為與ConvNets更加相似。Swin Transformer是這個方向上的一個里程碑式的工作，首次證明了Transformer可以作為通用的視覺骨干，并在圖像分類之外的一系列計算機視覺任務中取得最先進的性能。Swin Transformer的成功和快速采用也揭示了一件事：卷積的本質并沒有變得不重要；相反，它仍然備受期待，從未褪色。

Under this perspective, many of the advancements of Transformers for computer vision have been aimed at bringing back convolutions. These attempts, however, come at a cost: a naive implementation of sliding window self-attention can be expensive [55]; with advanced approaches such as cyclic shifting [45], the speed can be optimized but the system becomes more sophisticated in design. On the other hand, it is almost ironic that a ConvNet already satisfies many of those desired properties, albeit in a straightforward, no-frills way. The only reason ConvNets appear to be losing steam is that (hierarchical) Transformers surpass them in many vision tasks, and the performance difference is usually attributed to the superior scaling behavior of Transformers, with multi-head self-attention being the key component.

在這種觀點下，許多用于計算機視覺的Transformer的進步都是為了讓卷積回歸。然而，這些嘗試是有代價的：樸實的滑動窗口self-attention的實現可能是昂貴的；用先進的方法，如循環移位（cyclic shifting），速度可以被優化，但系統的設計變得更加復雜。另一方面，具有諷刺意味的是，ConvNet已經滿足了許多這些期望的特性，盡管是以一種直接的、不加修飾的方式。ConvNets似乎正在失去動力的唯一原因是（分層的）Transformers在許多視覺任務中超過了它們，而性能差異通常歸因于Transformers卓越的擴展行為，其中多頭自注意力是關鍵的組成部分。

Unlike ConvNets, which have progressively improved over the last decade, the adoption of Vision Transformers was a step change. In recent literature, system-level comparisons (e.g. a Swin Transformer vs. a ResNet) are usually adopted when comparing the two. ConvNets and hierarchical vision Transformers become different and similar at the same time: they are both equipped with similar inductive biases, but differ significantly in the training procedure and macro/micro-level architecture design. In this work, we investigate the architectural distinctions between ConvNets and Transformers and try to identify the confounding variables when comparing the network performance. Our research is intended to bridge the gap between the pre-ViT and post-ViT eras for ConvNets, as well as to test the limits of what a pure ConvNet can achieve.

與ConvNets不同的是，在過去的十年中，ConvNets逐步得到了改善，而采用Vision Transformers則是一個步驟的改變。在最近的文獻中，在比較兩者時通常采用系統級的比較（如Swin Transformer與ResNet）。ConvNets和分層視覺Transformer同時變得既不同又相似：它們都配備了類似的歸納偏置，但在訓練程序和宏觀/微觀層面的架構設計上有很大的不同。在這項工作中，我們研究了ConvNets和Transformers之間的架構區別，并試圖確定比較網絡性能時的混雜變量。我們的研究旨在彌合ConvNets的前ViT時代和后ViT時代之間的差距，以及測試純ConvNet能夠實現的極限。

To do this, we start with a standard ResNet (e.g. ResNet50) trained with an improved procedure. We gradually “modernize” the architecture to the construction of a hierarchical vision Transformer (e.g. Swin-T). Our exploration is directed by a key question: How do design decisions in Transformers impact ConvNets’ performance? We discover several key components that contribute to the performance difference along the way. As a result, we propose a family of pure ConvNets dubbed ConvNeXt. We evaluate ConvNeXts on a variety of vision tasks such as ImageNet classification [17], object detection/segmentation on COCO [44], and semantic segmentation on ADE20K [92]. Surprisingly, ConvNeXts, constructed entirely from standard ConvNet modules, compete favorably with Transformers in terms of accuracy, scalability and robustness across all major benchmarks. ConvNeXt maintains the efficiency of standard ConvNets, and the fully-convolutional nature for both training and testing makes it extremely simple to implement.

為了做到這一點，我們從一個標準的ResNet（例如ResNet50）開始，用改進的程序進行訓練。我們逐漸將架構 “現代化”，以構建一個分層的視覺Transformer（例如Swin-T）。我們的探索是由一個關鍵問題引導的。Transformer中的設計決定如何影響ConvNets的性能？我們發現了幾個關鍵的組件，這些組件有助于沿途的性能差異。因此，我們提出了一個被稱為ConvNeXt的純ConvNets系列。我們在各種視覺任務上評估了ConvNeXts，如ImageNet分類、COCO上的物體檢測/分割，以及ADE20K上的語義分割。令人驚訝的是，完全由標準ConvNet模塊構建的ConvNeXts在所有主要基準的準確性、可擴展性和魯棒性方面與Transformers競爭。ConvNeXt保持了標準ConvNets的效率，而且訓練和測試的完全卷積性質使其實現起來非常簡單。

We hope the new observations and discussions can challenge some common beliefs and encourage people to rethink the importance of convolutions in computer vision.

我們希望新的觀察和討論可以挑戰一些常見的信念，鼓勵人們重新思考計算機視覺中卷積的重要性。

2. Modernizing a ConvNet: a Roadmap —— 現代化的ConvNet：一個路線圖

In this section, we provide a trajectory going from a ResNet to a ConvNet that bears a resemblance to Transformers. We consider two model sizes in terms of FLOPs, one is the ResNet-50 / Swin-T regime with FLOPs around 4.5×109 and the other being ResNet-200 / Swin-B regime which has FLOPs around 15.0 × 109. For simplicity, we will present the results with the ResNet-50 / Swin-T complexity models. The conclusions for higher capacity models are consistent and results can be found in Appendix C.

在這一節中，我們提供了一個從ResNet到ConvNet的軌跡，這個軌跡與Transformer很相似。我們考慮了兩種FLOPs大小的模型，一種是ResNet-50 / Swin-T制度，FLOPs約為 $4.5×1094.5\times 10^9$ ，另一種是ResNet-200 / Swin-B制度，FLOPs約為 $15.0×10915.0\times 10^9$ 。為了簡單起見，我們將介紹ResNet-50 / Swin-T復雜度模型的結果。更高容量模型的結論是一致的，結果可以在附錄C中找到。

At a high level, our explorations are directed to investigate and follow different levels of designs from a Swin Transformer while maintaining the network’s simplicity as a standard ConvNet. The roadmap of our exploration is as follows. Our starting point is a ResNet-50 model. We first train it with similar training techniques used to train vision Transformers and obtain much improved results compared to the original ResNet-50. This will be our baseline. We then study a series of design decisions which we summarized as 1) macro design, 2) ResNeXt, 3) inverted bottleneck, 4) large kernel size, and 5) various layer-wise micro designs. In Figure 2, we show the procedure and the results we are able to achieve with each step of the “network modernization”. Since network complexity is closely correlated with the final performance, the FLOPs are roughly controlled over the course of the exploration, though at intermediate steps the FLOPs might be higher or lower than the reference models. All models are trained and evaluated on ImageNet-1K.

在高層(high level)上，我們的探索方向是研究和遵循Swin Transformer的不同層次(level)的設計，同時保持網絡作為一個標準ConvNet的簡單性。我們探索的路線圖如下。

Figure 2. We modernize a standard ConvNet (ResNet) towards the design of a hierarchical vision Transformer (Swin), without introducing any attention-based modules. The foreground bars are model accuracies in the ResNet-50/Swin-T FLOP regime; results for the ResNet-200/Swin-B regime are shown with the gray bars. A hatched bar means the modification is not adopted. Detailed results for both regimes are in the appendix. Many Transformer architectural choices can be incorporated in a ConvNet, and they lead to increasingly better performance. In the end, our pure ConvNet model, named ConvNeXt, can outperform the Swin Transformer.
圖2. 我們將一個標準的ConvNet（ResNet）現代化，以設計一個層次化的視覺Transformer（Swin），而不引入任何基于注意力的模塊。前面的條形圖是ResNet-50/Swin-T FLOP體系中的模型精度；ResNet-200/Swin-B體系的結果用灰色條形圖表示。帶帽子的條形圖表示沒有采用該修改。兩個制度的詳細結果見附錄。許多Transformer架構的選擇可以被納入ConvNet中，而且它們會帶來越來越好的性能。最后，我們的純ConvNet模型，名為ConvNeXt，可以超過Swin Transformer。

我們的起點是一個ResNet-50模型。我們首先用類似于訓練視覺Transformer的訓練技巧來訓練它，并獲得比原來的ResNet-50更多的結果。這將是我們的基線（baseline）。然后，我們研究了一系列的設計決策，我們總結為：

宏觀設計

ResNeXt

倒置瓶頸

大卷積核

各種層級的微設計

在圖2中，我們展示了 "網絡現代化 "的每一步的程序和我們能夠實現的結果。由于網絡的復雜性與最終的性能密切相關，在探索的過程中，FLOPs被大致控制，盡管在中間步驟，FLOPs可能高于或低于參考模型。所有模型都是在ImageNet-1K上訓練和評估的。

2.1 Training Techniques —— 訓練技巧

Apart from the design of the network architecture, the training procedure also affects the ultimate performance. Not only did vision Transformers bring a new set of modules and architectural design decisions, but they also introduced different training techniques (e.g. AdamW optimizer) to vision. This pertains mostly to the optimization strategy and associated hyper-parameter settings. Thus, the first step of our exploration is to train a baseline model with the vision Transformer training procedure, in this case, ResNet50/200. Recent studies [7, 81] demonstrate that a set of modern training techniques can significantly enhance the performance of a simple ResNet-50 model. In our study, we use a training recipe that is close to DeiT’s [73] and Swin Transformer’s [45]. The training is extended to 300 epochs from the original 90 epochs for ResNets. We use the AdamW optimizer [46], data augmentation techniques such as Mixup [90], Cutmix [89], RandAugment [14], Random Erasing [91], and regularization schemes including Stochastic Depth [36] and Label Smoothing [69]. The complete set of hyper-parameters we use can be found in Appendix A.1. By itself, this enhanced training recipe increased the performance of the ResNet-50 model from 76.1% [1] to 78.8% (+2.7%), implying that a significant portion of the performance difference between traditional ConvNets and vision Transformers may be due to the training techniques. We will use this fixed training recipe with the same hyperparameters throughout the “modernization” process. Each reported accuracy on the ResNet-50 regime is an average obtained from training with three different random seeds.

除了網絡架構的設計，訓練程序也會影響最終的性能。視覺Transformer不僅帶來了一套新的模塊和架構設計決策，而且還為視覺引入了不同的訓練技術（如AdamW優化器）。這主要涉及到優化策略和相關的超參數設置。因此，我們探索的第一步是用視覺Transformer訓練程序訓練一個基線模型（baseline），在這種情況下是ResNet50/200。最近的研究表明，一套現代訓練技術可以顯著提高一個簡單的ResNet-50模型的性能。在我們的研究中，我們使用了與DeiT和Swin Transformer的相近的訓練配置。訓練從原來的90個epochs擴展到300個epochs的ResNets。我們使用AdamW優化器，數據增強技術，如Mixup、Cutmix、RandAugment、Random Erasing，以及包括Stochastic Depth和Label Smoothing的正則化方案。我們使用的完整的超參數集可以在附錄A.1中找到。

就其本身而言，這個增強的訓練配置將ResNet-50模型的性能從76.1%提高到78.8%（+2.7%），這意味著傳統ConvNets和視覺Transformer之間的性能差異的很大一部分可能是由于訓練技巧造成的。我們將在整個 "現代化"過程中使用這個固定的訓練配置，并使用相同的超參數。ResNet-50制度上的每個報告的準確度是用三個不同的隨機種子訓練得到的平均值。

2.2 Macro Design —— 宏觀設計

We now analyze Swin Transformers’ macro network design. Swin Transformers follow ConvNets [28, 65] to use a multi-stage design, where each stage has a different feature map resolution. There are two interesting design considerations: the stage compute ratio, and the “stem cell” structure.

我們現在分析一下Swin Transformers的宏觀網絡設計。Swin Transformers跟隨ConvNets使用多階段設計，每個階段有不同的特征圖分辨率。有兩個有趣的設計考慮：階段計算比和 "干細胞(stem cell)"結構。

Changing stage compute ratio. The original design of the computation distribution across stages in ResNet was largely empirical. The heavy “res4” stage was meant to be compatible with downstream tasks like object detection, where a detector head operates on the 14×14 feature plane. Swin-T, on the other hand, followed the same principle but with a slightly different stage compute ratio of 1:1:3:1. For larger Swin Transformers, the ratio is 1:1:9:1. Following the design, we adjust the number of blocks in each stage from (3, 4, 6, 3) in ResNet-50 to (3, 3, 9, 3), which also aligns the FLOPs with Swin-T. This improves the model accuracy from 78.8% to 79.4%. Notably, researchers have thoroughly investigated the distribution of computation [53, 54], and a more optimal design is likely to exist.
From now on, we will use this stage compute ratio.

2.2.1 改變階段性的計算比例 (Changing stage compute ratio)

ResNet中各階段的計算分布的最初設計主要是經驗性的。沉重的 "res4 "階段是為了與下游任務兼容，如目標檢測，其中一個檢測器頭（detector head）在14×14的特征平面上操作。另一方面，Swin-T也遵循同樣的原則，但階段計算比例略有不同，為1：1：3：1。對于較大的Swin Transformers，比例為1：1：9：1。按照設計，我們將每個階段的塊數(blocks)從ResNet-50的（3，4，6，3）調整為（3，3，9，3），這也使FLOPs與Swin-T一致。這使模型的準確性從78.8%提高到79.4%。值得注意的是，研究人員已經徹底調查了計算的分布情況，而且很可能存在一個更理想的設計。

從現在開始，我們將使用這個階段的計算比例。

Changing stem to “Patchify”. Typically, the stem cell design is concerned with how the input images will be processed at the network’s beginning. Due to the redundancy inherent in natural images, a common stem cell will aggressively downsample the input images to an appropriate feature map size in both standard ConvNets and vision Transformers. The stem cell in standard ResNet contains a 7×7 convolution layer with stride 2, followed by a max pool, which results in a 4× downsampling of the input images. In vision Transformers, a more aggressive “patchify” strategy is used as the stem cell, which corresponds to a large kernel size (e.g. kernel size = 14 or 16) and non-overlapping convolution. Swin Transformer uses a similar “patchify” layer, but with a smaller patch size of 4 to accommodate the architecture’s multi-stage design. We replace the ResNet-style stem cell with a patchify layer implemented using a 4×4, stride 4 convolutional layer. The accuracy has changed from 79.4% to 79.5%. This suggests that the stem cell in a ResNet may be substituted with a simpler “patchify” layer à la ViT which will result in similar performance.
We will use the “patchify stem” (4×4 non-overlapping convolution) in the network.

2.2.2 將"stem"改為 “Patchify” (Changing stem to “Patchify”)。

通常情況下，stem設計關注的是在網絡開始時如何處理輸入圖像。由于自然圖像中固有的冗余，一個普通的stem層將積極地對輸入圖像進行降采樣，以達到標準卷積網絡和視覺Transformer中適當的特征圖大小。標準ResNet中的stem層包含一個7×7的卷積層，步長為2，然后是一個MaxPooling層，這導致輸入圖像的4倍下采樣。在視覺Transformer中，一個更激進的 "Patchify"策略被用作stem層，它對應于一個大的核大小（例如kernel size=14或16）和非重疊卷積。Swin Transformer使用類似的 "Patchify "層，但patch尺寸較小，為4，以適應架構的多階段設計。我們用一個使用4×4、步長為4的卷積層實現的patchify層取代ResNet式的stem層。準確率從79.4%變為79.5%。這表明ResNet中的stem層可以用一個更簡單的 "patchify "層來代替，就像ViT一樣，這將導致類似的性能。

我們將在網絡中使用 “patchify stem”（4×4非重疊卷積）。

非重疊卷積就是說卷積核大小 $≤\le$ 步長

2.3. ResNeXt-ify —— ResNeXT化

In this part, we attempt to adopt the idea of ResNeXt [87], which has a better FLOPs/accuracy trade-off than a vanilla ResNet. The core component is grouped convolution, where the convolutional filters are separated into different groups. At a high level, ResNeXt’s guiding principle is to “use more groups, expand width”. More precisely, ResNeXt employs grouped convolution for the 3×3 conv layer in a bottleneck block. As this significantly reduces the FLOPs, the network width is expanded to compensate for the capacity loss.

在這一部分，我們試圖采用ResNeXt的思想，它比普通的ResNet有更好的FLOPs/準確性權衡。其核心部分是分組卷積，其中卷積卷積核被分成不同的組。在高層次上，ResNeXt的指導原則是 “使用更多的組，擴大寬度”。更確切地說，ResNeXt對Bottleneck中的3×3卷積層采用了分組卷積。由于這大大減少了FLOPs，網絡寬度被擴大以補償容量的損失。

In our case we use depthwise convolution, a special case of grouped convolution where the number of groups equals the number of channels. Depthwise conv has been popularized by MobileNet [34] and Xception [11]. We note that depthwise convolution is similar to the weighted sum operation in self-attention, which operates on a per-channel basis, i.e., only mixing information in the spatial dimension. The combination of depthwise conv and 1 × 1 convs leads to a separation of spatial and channel mixing, a property shared by vision Transformers, where each operation either mixes information across spatial or channel dimension, but not both. The use of depthwise convolution effectively reduces the network FLOPs and, as expected, the accuracy. Following the strategy proposed in ResNeXt, we increase the network width to the same number of channels as Swin-T’s (from 64 to 96). This brings the network performance to 80.5% with increased FLOPs (5.3G). We will now employ the ResNeXt design.

在我們的案例中，我們使用深度卷積，這是分組卷積的一個特例，其中分組的數量等于通道的數量。深度卷積已被MobileNet和Xception所推廣。我們注意到，深度卷積與自注意中的加權和操作類似，后者是在每個通道的基礎上操作的，也就是說，只混合空間維度的信息。深度卷積和1×1卷積的結合導致了空間和通道混合的分離，這是視覺Transformer所共有的屬性，每個操作要么在空間或通道維度上混合信息，但不能同時混合。深度卷積的使用有效地減少了網絡的FLOPs，正如預期的那樣，也減少了準確性。按照ResNeXt提出的策略，我們將網絡寬度增加到與Swin-T的通道數量相同（從64到96）。這使得網絡性能達到80.5%，FLOPs增加（5.3G）。我們現在將采用ResNeXt的設計。

2.4. Inverted Bottleneck —— 逆殘差模塊

One important design in every Transformer block is that it creates an inverted bottleneck, i.e., the hidden dimension of the MLP block is four times wider than the input dimension (see Figure 4). Interestingly, this Transformer design is connected to the inverted bottleneck design with an expansion ratio of 4 used in ConvNets. The idea was popularized by MobileNetV2 [61], and has subsequently gained traction in several advanced ConvNet architectures [70, 71].

每個Transformer塊中的一個重要設計是，它創造了一個逆殘差瓶頸模塊，即MLP塊的隱藏維度比輸入維度寬四倍（見圖4）。有趣的是，這種Transformer設計與ConvNets中使用的擴展率為4的逆殘差瓶頸模塊設計有聯系。這個想法被MobileNetV2所推廣，隨后在一些先進的ConvNet架構中得到推廣[MnasNet, EfficientNet]。

Figure 3. Block modifications and resulted specifications. (a) is a ResNeXt block; in (b) we create an inverted bottleneck block and in ? the position of the spatial depthwise conv layer is moved up.
圖3. 塊的修改和結果規格。(a)是一個ResNeXt塊；在(b)中，我們創建了一個倒置的瓶頸塊，在?中，空間縱深說服層的位置被上移。

Here we explore the inverted bottleneck design. Figure 3 (a) to (b) illustrate the configurations. Despite the increased FLOPs for the depthwise convolution layer, this change reduces the whole network FLOPs to 4.6G, due to the significant FLOPs reduction in the downsampling residual blocks’ shortcut 1×1 conv layer. Interestingly, this results in slightly improved performance (80.5% to 80.6%). In the ResNet-200 / Swin-B regime, this step brings even more gain (81.9% to 82.6%) also with reduced FLOPs.
We will now use inverted bottlenecks.

在這里，我們探討了逆殘差模塊設計。圖3（a）至（b）說明了配置。盡管深度卷積層的FLOPs增加了，但由于下采樣殘余塊的捷徑1×1卷積層的FLOPs大幅減少，這種改變使整個網絡的FLOPs減少到4.6G。有趣的是，這樣做的結果是性能略有提高（80.5%到80.6%）。在ResNet-200/Swin-B系統中，這一步帶來了更多的收益（81.9%到82.6%），也減少了FLOPs。

我們現在將使用逆殘差模塊。

2.5. Large Kernel Sizes —— 大卷積核

In this part of the exploration, we focus on the behavior of large convolutional kernels. One of the most distinguishing aspects of vision Transformers is their non-local self-attention, which enables each layer to have a global receptive field. While large kernel sizes have been used in the past with ConvNets [40, 68], the gold standard (popularized by VGGNet [65]) is to stack small kernel-sized (3×3) conv layers, which have efficient hardware implementations on modern GPUs [41]. Although Swin Transformers reintroduced the local window to the self-attention block, the window size is at least 7×7, significantly larger than the ResNe(X)t kernel size of 3×3. Here we revisit the use of large kernel-sized convolutions for ConvNets.

在這一部分的探索中，我們重點關注大型卷積核的效果。視覺Transformer最突出的一個方面是它們的非局部自我注意（non-local self-attention），這使得每一層都有一個全局的接受場（global receptive field）。雖然過去在ConvNets[AlexNet, Inception v1]中使用了大內核尺寸，但黃金標準（由VGGNet推廣）是堆疊小內核尺寸（3×3）的conv層，這在現代GPU上有高效的硬件實現。雖然Swin Transformers在自注意力模塊中重新引入了局部窗口（local window），但窗口大小至少是7×7，明顯大于3×3的ResNe(X)t內核大小。在此，我們重新審視大核大小的卷積在ConvNets中的使用。

2.5.1 Moving up depthwise conv layer —— 上移深度卷積層

To explore large kernels, one prerequisite is to move up the position of the depthwise conv layer (Figure 3 (b) to ?). That is a design decision also evident in Transformers: the MSA block is placed prior to the MLP layers. As we have an inverted bottleneck block, this is a natural design choice — the complex/inefficient modules (MSA, large-kernel conv) will have fewer channels, while the efficient, dense 1×1 layers will do the heavy lifting. This intermediate step reduces the FLOPs to 4.1G, resulting in a temporary performance degradation to 79.9%.

為了探索大的內核，一個前提條件是將深度卷積層的位置上移（圖3（b）到（c））。這是一個在Transformer中也很明顯的設計決定：MSA塊被放在MLP層之前。由于我們有一個逆殘差模塊，這是一個自然的設計選擇——復雜/低效的模塊（MSA，大核conv）將有較少的通道，而高效、密集的1×1層將完成重任。這個中間步驟將FLOPs減少到4.1G，導致性能暫時下降到79.9%。

2.5.2 Increasing the kernel size —— 增大卷積核尺寸

With all of these preparations,the benefit of adopting larger kernel-sized convolutions is significant. We experimented with several kernel sizes, including 3, 5, 7, 9, and 11. The network’s performance increases from 79.9% (3×3) to 80.6% (7×7), while the network’s FLOPs stay roughly the same. Additionally, we observe that the benefit of larger kernel sizes reaches a saturation point at 7×7. We verified this behavior in the large capacity model too: a ResNet-200 regime model does not exhibit further gain when we increase the kernel size beyond 7×7.
We will use 7×7 depthwise conv in each block.
At this point, we have concluded our examination of network architectures on a macro scale. Intriguingly, a significant portion of the design choices taken in a vision Transformer may be mapped to ConvNet instantiations.

在所有這些準備工作中，采用較大的核大小的卷積的好處是顯著的。我們試驗了幾種內核大小，包括3、5、7、9和11。網絡的性能從79.9%（3×3）增加到80.6%（7×7），而網絡的FLOPs大致保持不變。此外，我們觀察到，更大的內核尺寸的好處在7×7時達到了飽和點。我們在大容量模型中也驗證了這種行為：當我們將核大小增加到7×7以上時，ResNet-200制度模型沒有表現出進一步的收益。

我們將在每個區塊中使用7×7的深度 conv。

至此，我們結束了對宏觀規模上的網絡結構的研究。耐人尋味的是，在視覺Transformer中采取的相當一部分設計選擇可以映射到ConvNet實例中。

2.6. Micro Design —— 微觀設計

In this section, we investigate several other architectural differences at a micro scale — most of the explorations here are done at the layer level, focusing on specific choices of activation functions and normalization layers.

在本節中，我們在微觀層面上研究了其他幾個架構上的差異——這里的大部分探索都是在層級上完成的，重點是激活函數和歸一化層的具體選擇。

2.6.1 Replacing ReLU with GELU

One discrepancy between NLP and vision architectures is the specifics of which activation functions to use. Numerous activation functions have been developed over time, but the Rectified Linear Unit (ReLU) [49] is still extensively used in ConvNets due to its simplicity and efficiency. ReLU is also used as an activation function in the original Transformer paper [77]. The Gaussian Error Linear Unit, or GELU [32], which can be thought of as a smoother variant of ReLU, is utilized in the most advanced Transformers, including Google’s BERT [18] and OpenAI’s GPT-2 [52], and, most recently, ViTs. We find that ReLU can be substituted with GELU in our ConvNet too, although the accuracy stays unchanged (80.6%).

NLP和視覺架構之間的一個差異是使用何種激活函數的具體問題。隨著時間的推移，許多激活函數已經被開發出來，但整流線性單元（ReLU）由于其簡單和高效，仍然被廣泛用于ConvNets。ReLU也被用作原始變形器論文中的激活函數。高斯誤差線性單元，即GELU，可以被認為是ReLU的平滑變體，在最先進的Transformer中被利用，包括谷歌的BERT和OpenAI的GPT-2，以及最近的ViTs。我們發現，在我們的ConvNet中，ReLU也可以用GELU代替，盡管準確率保持不變（80.6%）。

2.6.2 Fewer activation functions —— 更少的激活函數

One minor distinction between a Transformer and a ResNet block is that Transformers have fewer activation functions. Consider a Transformer block with key/query/value linear embedding layers, the projection layer, and two linear layers in an MLP block. There is only one activation function present in the MLP block. In comparison, it is common practice to append an activation function to each convolutional layer, including the 1 × 1 convs. Here we examine how performance changes when we stick to the same strategy. As depicted in Figure 4, we eliminate all GELU layers from the residual block except for one between two 1 × 1 layers, replicating the style of a Transformer block. This process improves the result by 0.7% to 81.3%, practically matching the performance of Swin-T.
We will now use a single GELU activation in each block.

Transformer和ResNet塊之間的一個小區別是，Transformer的激活函數較少。考慮一個帶有鍵（Key）/查詢（Query）/值（Value）線性嵌入層（Embedding層）的Transformer塊，投影層（projection layer），以及MLP塊中的兩個線性層（linear layer）。在MLP塊中只有一個激活函數存在。相比之下，通常的做法是在每個卷積層（包括1×1卷積層）上附加一個激活函數。在這里，我們研究了當我們堅持使用相同的策略時，性能如何變化，如圖4所示。

Figure 4. Block designs for a ResNet, a Swin Transformer, and a ConvNeXt. Swin Transformer’s block is more sophisticated due to the presence of multiple specialized modules and two residual connections. For simplicity, we note the linear layers in Transformer MLP blocks also as “1×1 convs” since they are equivalent.
圖4. 一個ResNet、一個Swin Transformer和一個ConvNeXt的模塊設計。由于存在多個專門的模塊和兩個剩余連接，Swin Transformer的模塊更加復雜。為了簡單起見，我們把Transformer MLP塊中的線性層也記為 “1×1 convs”，因為它們是等同的。

我們從殘差塊中消除了所有的GELU層，除了兩個1×1層之間的一個，復制了變形塊的風格。這個過程將結果提高了0.7%，達到81.3%，實際上與Swin-T的性能相匹配。

現在我們將在每個塊中使用單一的GELU激活。

2.6.3 Fewer normalization layers —— 更少的歸一化層

Transformer blocks usually have fewer normalization layers as well. Here we remove two BatchNorm (BN) layers, leaving only one BN layer before the conv 1 × 1 layers. This further boosts the performance to 81.4%, already surpassing Swin-T’s result. Note that we have even fewer normalization layers per block than Transformers, as empirically we find that adding one additional BN layer at the beginning of the block does not improve the performance.

Transformer塊通常也有較少的歸一化層。這里我們去掉了兩個BatchNorm（BN）層，在Conv 1×1層之前只留下一個BN層。這進一步將性能提高到81.4%，已經超過了Swin-T的結果。請注意，我們每個區塊的歸一化層數甚至比Transformers還要少，因為根據經驗，我們發現在區塊的開始增加一個額外的BN層并不能提高性能。

2.6.4 Substituting BN with LN —— 使用LN替換BN

BatchNorm [38] is an essential component in ConvNets as it improves the convergence and reduces overfitting. However, BN also has many intricacies that can have a detrimental effect on the model’s performance [84]. There have been numerous attempts at developing alternative normalization [60, 75, 83] techniques, but BN has remained the preferred option in most vision tasks. On the other hand, the simpler Layer Normalization [5] (LN) has been used in Transformers, resulting in good performance across different application scenarios.

BatchNorm是ConvNets中的一個重要組成部分，因為它可以提高收斂性并減少過擬合。然而，BN也有許多錯綜復雜的問題，會對模型的性能產生不利的影響[84]。已經有很多人嘗試開發替代的歸一化技術[60, 75, 83]，但在大多數視覺任務中，BN仍然是首選。另一方面，更簡單的層歸一化(LN)已被用于Transformer，在不同的應用場景中產生了良好的性能。

Directly substituting LN for BN in the original ResNet will result in suboptimal performance [83]. With all the modifications in network architecture and training techniques, here we revisit the impact of using LN in place of BN. We observe that our ConvNet model does not have any difficulties training with LN; in fact, the performance is slightly better, obtaining an accuracy of 81.5%.
From now on, we will use one LayerNorm as our choice of normalization in each residual block.

在原ResNet中直接用LN代替BN會導致次優的性能[83]。隨著網絡結構和訓練技術的所有修改，這里我們重新審視了使用LN來代替BN的影響。我們觀察到，我們的ConvNet模型在使用LN訓練時沒有任何困難；事實上，性能略好，獲得了81.5%的準確性。

從現在開始，我們將使用一個LayerNorm作為我們在每個殘差塊中的標準化選擇。

2.6.5 Separate downsampling layers獨立的下采樣層

In ResNet, the spatial downsampling is achieved by the residual block at the start of each stage, using 3×3 conv with stride 2 (and 1×1 conv with stride 2 at the shortcut connection). In Swin Transformers, a separate downsampling layer is added between stages. We explore a similar strategy in which we use 2×2 conv layers with stride 2 for spatial downsampling. This modification surprisingly leads to diverged training. Further investigation shows that, adding normalization layers wherever spatial resolution is changed can help stablize training. These include several LN layers also used in Swin Transformers: one before each downsampling layer, one after the stem, and one after the final global average pooling. We can improve the accuracy to 82.0%, significantly exceeding Swin-T’s 81.3%.

在ResNet中，空間下采樣是由每個階段開始時的residual block實現的，使用3×3 conv with stride 2（在捷徑連接處使用1×1 conv with stride 2）。在Swin Transformers中，在各階段之間增加了一個單獨的下采樣層。我們探索了一種類似的策略，即使用跨度為2的2×2 conv層進行空間下采樣。這種修改出人意料地導致了訓練的分歧。進一步的調查顯示，在空間分辨率改變的地方添加歸一化層，有助于穩定訓練。這些包括同樣用于Swin Transformers的幾個LN層：一個在每個下采樣層之前，一個在干層之后，一個在最后的全局平均匯集之后。我們可以將精度提高到82.0%，大大超過Swin-T的81.3%。

We will use separate downsampling layers. This brings us to our final model, which we have dubbed ConvNeXt. A comparison of ResNet, Swin, and ConvNeXt block structures can be found in Figure 4. A comparison of ResNet-50, Swin-T and ConvNeXt-T’s detailed architecture specifications can be found in Table 9.

我們將使用單獨的下采樣層。這給我們帶來了最終的模型，我們將其稱為ConvNeXt。圖4是ResNet、Swin和ConvNeXt塊結構的比較。ResNet-50、Swin-T和ConvNeXt-T的詳細架構規格的比較可以在表9中找到。

2.6.6 Closing remarks —— 閉幕詞（總結）

We have finished our first “playthrough” and discovered ConvNeXt, a pure ConvNet, that can outperform the Swin Transformer for ImageNet-1K classification in this compute regime. It is worth noting that all design choices discussed so far are adapted from vision Transformers. In addition, these designs are not novel even in the ConvNet literature — they have all been researched separately, but not collectively, over the last decade. Our ConvNeXt model has approximately the same FLOPs, #params., throughput, and memory use as the Swin Transformer, but does not require specialized modules such as shifted window attention or relative position biases.

我們已經完成了我們的第一周目，發現ConvNeXt，一個純粹的ConvNet，在這個計算系統中可以超過Swin Transformer的ImageNet-1K分類。值得注意的是，到目前為止討論的所有設計選擇都是從視覺變形器中改編而來。此外，這些設計即使在ConvNet文獻中也并不新穎——它們在過去十年中都被單獨研究過，但沒有被集體研究過。我們的ConvNeXt模型的FLOPs、#params.、吞吐量和內存使用量與Swin Transformer大致相同，但不需要專門的模塊，如移窗注意或相對位置偏差。

These findings are encouraging but not yet completely convincing — our exploration thus far has been limited to a small scale, but vision Transformers’ scaling behavior is what truly distinguishes them. Additionally, the question of whether a ConvNet can compete with Swin Transformers on downstream tasks such as object detection and semantic segmentation is a central concern for computer vision practitioners. In the next section, we will scale up our ConvNeXt models both in terms of data and model size, and evaluate them on a diverse set of visual recognition tasks.

這些發現令人鼓舞，但還不能完全令人信服——迄今為止，我們的探索僅限于小規模，但視覺Transformer的擴展行為才是它們真正的區別所在。此外，ConvNet能否在下游任務（如物體檢測和語義分割）上與Swin Transformers競爭的問題是計算機視覺從業者的核心關注點。在下一節中，我們將在數據和模型大小方面擴大我們的ConvNeXt模型，并在一組不同的視覺識別任務上對它們進行評估。

3. Empirical Evaluations on ImageNet —— 在ImageNet上的經驗評估

We construct different ConvNeXt variants, ConvNeXtT/S/B/L, to be of similar complexities to Swin-T/S/B/L [45]. ConvNeXt-T/B is the end product of the “modernizing” procedure on ResNet-50/200 regime, respectively. In addition, we build a larger ConvNeXt-XL to further test the scalability of ConvNeXt. The variants only differ in the number of channels C, and the number of blocks B in each stage. Following both ResNets and Swin Transformers, the number of channels doubles at each new stage. We summarize the configurations below:

我們構建了不同的ConvNeXt變體，ConvNeXtT/S/B/L，其復雜程度與Swin-T/S/B/L相似。ConvNeXt-T/B是在ResNet-50/200制度上分別進行 "現代化 "程序的最終產品。此外，我們建立了一個更大的ConvNeXt-XL來進一步測試ConvNeXt的可擴展性。這些變體只在通道數C和每個階段的塊數B上有所不同。按照ResNets和Swin Transformers，通道的數量在每個新階段都會增加一倍。我們把這些配置總結如下。

? ConvNeXt-T: C = (96, 192, 384, 768), B = (3, 3, 9, 3)
? ConvNeXt-S: C = (96, 192, 384, 768), B = (3, 3, 27, 3)
? ConvNeXt-B: C = (128, 256, 512, 1024), B = (3, 3, 27, 3)
? ConvNeXt-L: C = (192, 384, 768, 1536), B = (3, 3, 27, 3)
? ConvNeXt-XL: C = (256, 512, 1024, 2048), B = (3, 3, 27, 3)

3.1. Settings

The ImageNet-1K dataset consists of 1000 object classes with 1.2M training images. We report ImageNet-1K top-1 accuracy on the validation set. We also conduct pre-training on ImageNet-22K, a larger dataset of 21841 classes (a superset of the 1000 ImageNet-1K classes) with ～14M images for pre-training, and then fine-tune the pre-trained model on ImageNet-1K for evaluation. We summarize our training setups below. More details can be found in Appendix A.

ImageNet-1K數據集由1000個物體類別和120萬張訓練圖像組成。我們報告了ImageNet-1K在驗證集上的最高準確性。我們還在ImageNet-22K上進行了預訓練，這是一個由21841個類組成的更大的數據集（1000個ImageNet-1K類的超集），有1400萬張圖像用于預訓練，然后在ImageNet-1K上對預訓練模型進行微調以進行評估。我們在下面總結了我們的訓練設置。更多的細節可以在附錄A中找到。

Training on ImageNet-1K

We train ConvNeXts for 300 epochs using AdamW [46] with a learning rate of 4e-3. There is a 20-epoch linear warmup and a cosine decaying schedule afterward. We use a batch size of 4096 and a weight decay of 0.05. For data augmentations, we adopt common schemes including Mixup [90], Cutmix [89], RandAugment [14], and Random Erasing [91]. We regularize the networks with Stochastic Depth [37] and Label Smoothing [69]. Layer Scale [74] of initial value 1e-6 is applied. We use Exponential Moving Average (EMA) [51] as we find it alleviates larger models’ overfitting.

我們使用AdamW對ConvNeXts進行了300個epochs的訓練，學習率為 $4×10?34\times 10^{-3}$ 。有一個20個epoch的線性預熱，之后是余弦衰落的時間表。我們使用了4096的批次大小和0.05的權重衰減。對于數據增強，我們采用常見的方案，包括Mixup、Cutmix、RandAugment和Random Erasing。我們用隨機深度和標簽平滑對網絡進行規范。采用了初始值為 $1×10?61\times 10^{-6}$ 的Layer Scale。我們使用指數移動平均法（EMA），因為我們發現它可以減輕較大的模型的過擬合。

Pre-training on ImageNet-22K

We pre-train ConvNeXts on ImageNet-22K for 90 epochs with a warmup of 5 epochs. We do not use EMA. Other settings follow ImageNet-1K.

我們在ImageNet-22K上對ConvNeXts進行了90個epochs的預訓練，并進行了5個epochs的預熱。我們不使用EMA。其他設置遵循ImageNet-1K。

Fine-tuning on ImageNet-1K

We fine-tune ImageNet-22K pre-trained models on ImageNet-1K for 30 epochs. We use AdamW, a learning rate of 5e-5, cosine learning rate schedule, layer-wise learning rate decay [6, 12], no warmup, a batch size of 512, and weight decay of 1e-8. The default pre-training, fine-tuning, and testing resolution is 2242 . Additionally, we fine-tune at a larger resolution of 3842, for both ImageNet-22K and ImageNet-1K pre-trained models.

我們在ImageNet-1K上對ImageNet-22K的預訓練模型進行了30個epochs的微調。我們使用AdamW，學習率為 $5×10?55\times 10^{-5}$ ，余弦學習率計劃，層級學習率衰減[6, 12]，無預熱，批次大小為512，權重衰減為 $1×10?81\times 10^{-8}$ 。默認的預訓練、微調和測試分辨率為 $224^2$ 。此外，我們對ImageNet-22K和ImageNet-1K的預訓練模型在更大的分辨率下進行微調，即 $384^2$ 。

Compared with ViTs/Swin Transformers, ConvNeXts are simpler to fine-tune at different resolutions, as the network is fully-convolutional and there is no need to adjust the input patch size or interpolate absolute/relative position biases.

與ViTs/Swin Transformers相比，ConvNeXts在不同分辨率下的微調更簡單，因為網絡是完全卷積的，不需要調整輸入補丁大小或插值絕對/相對位置偏差。

3.2. Results

ImageNet-1K

Table 1 (upper) shows the result comparison with two recent Transformer variants, DeiT [73] and Swin Transformers [45], as well as two ConvNets from architecture search - RegNets [54], EfficientNets [71] and EfficientNetsV2 [72]. ConvNeXt competes favorably with two strong ConvNet baselines (RegNet [54] and EfficientNet [71]) in terms of the accuracy-computation trade-off, as well as the inference throughputs. ConvNeXt also outperforms Swin Transformer of similar complexities across the board, sometimes with a substantial margin (e.g. 0.8% for ConvNeXt-T). Without specialized modules such as shifted windows or relative position bias, ConvNeXts also enjoy improved throughput compared to Swin Transformers.

表1（上）顯示了與最近的兩個Transformer變體DeiT和Swin Transformers，以及兩個來自架構搜索的ConvNets–RegNets、EfficientNets和EfficientNetsV2的結果比較。

Table 1. Classification accuracy on ImageNet-1K. Similar to Transformers, ConvNeXt also shows promising scaling behavior with higher-capacity models and a larger (pre-training) dataset. Inference throughput is measured on a V100 GPU, following [45]. On an A100 GPU, ConvNeXt can have a much higher throughput than Swin Transformer. See Appendix E. (?)ViT results with 90-epoch AugReg [67] training, provided through personal communication with the authors.
表1. ImageNet-1K的分類精度。與Transformers類似，ConvNeXt也顯示了在更高容量的模型和更大的（預訓練）數據集下有希望的擴展行為。推理吞吐量是在V100 GPU上測量的，遵循[Swin-Transformer]。在A100 GPU上，ConvNeXt的吞吐量可以比Swin Transformer高得多。見附錄E。(?)ViT在90個周期的AugReg[67]訓練下的結果，通過與作者的個人交流提供。

ConvNeXt與兩個強大的ConvNet基線（RegNet和EfficientNet）在準確性-計算權衡以及推理吞吐量方面進行了良好的競爭。ConvNeXt也全面超越了復雜程度相似的Swin Transformer，有時還有很大的差距（例如ConvNeXt-T的0.8%）。如果沒有專門的模塊，如移位窗口(shifted windows)或相對位置偏差(relative position bias)，ConvNeXt也享有比Swin Transformer更好的吞吐量(throughput )。

A highlight from the results is ConvNeXt-B at $384^2$ : it outperforms Swin-B by 0.6% (85.1% vs. 84.5%), but with 12.5% higher inference throughput (95.7 vs. 85.1 image/s). We note that the FLOPs/throughput advantage of ConvNeXt-B over Swin-B becomes larger when the resolution increases from $224^2$ to $384^2$ . Additionally, we observe an improved result of 85.5% when further scaling to ConvNeXt-L.

結果中的一個亮點是 $384^2$ 的ConvNeXt-B：它比Swin-B高出0.6%（85.1%對84.5%），但推理吞吐量高出12.5%（95.7對85.1圖像/秒）。我們注意到，當分辨率從 $224^2$ 增加到 $384^2$ 時，ConvNeXt-B相對于Swin-B的FLOPs/吞吐量優勢變得更大。此外，當進一步擴展到ConvNeXt-L時，我們觀察到85.5%的改進結果。

ImageNet-22K. We present results with models fine-tuned from ImageNet-22K pre-training at Table 1 (lower). These experiments are important since a widely held view is that vision Transformers have fewer inductive biases thus can perform better than ConvNets when pre-trained on a larger scale. Our results demonstrate that properly designed ConvNets are not inferior to vision Transformers when pre-trained with large dataset — ConvNeXts still perform on par or better than similarly-sized Swin Transformers, with slightly higher throughput. Additionally, our ConvNeXt-XL model achieves an accuracy of 87.8% — a decent improvement over ConvNeXt-L at $384^2$ , demonstrating that ConvNeXts are scalable architectures.

ImageNet-22K

我們在表1（下圖）展示了從ImageNet-22K預訓練中微調的模型結果。這些實驗是很重要的，因為有一種廣泛的觀點認為，視覺Transformer的歸納偏置較少，因此在進行大規模的預訓練時可以比ConvNets的表現更好。我們的結果表明，當用大型數據集進行預訓練時，適當設計的ConvNets并不遜于視覺Transformer——ConvNeXts的性能仍然與類似規模的Swin Transformers相當或更好，而且吞吐量略高。此外，我們的ConvNeXt-XL模型達到了87.8%的準確率——比ConvNeXt-L的 $384^2$ 的準確率有了很大的提高，這表明ConvNeXts是可擴展的架構。

On ImageNet-1K, EfficientNetV2-L, a searched architecture equipped with advanced modules (such as Squeeze-andExcitation [35]) and progressive training procedure achieves top performance. However, with ImageNet-22K pre-training, ConvNeXt is able to outperform EfficientNetV2, further demonstrating the importance of large-scale training.
In Appendix B, we discuss robustness and out-of-domain generalization results for ConvNeXt.

在ImageNet-1K上，EfficientNetV2-L，一個配備了高級模塊（如Squeeze-andExcitation[35]）和漸進式訓練程序的搜索架構取得了頂級性能。然而，在ImageNet-22K的預訓練下，ConvNeXt能夠超越EfficientNetV2，進一步證明了大規模訓練的重要性。

在附錄B中，我們討論了ConvNeXt的魯棒性（robustness）和域外泛化結果（out-of-domain generalization results）。

3.3. Isotropic ConvNeXt vs. ViT —— 各向同性研究

In this ablation, we examine if our ConvNeXt block design is generalizable to ViT-style [20] isotropic architectures which have no downsampling layers and keep the same feature resolutions (e.g. 14×14) at all depths. We construct isotropic ConvNeXt-S/B/L using the same feature dimensions as ViT-S/B/L (384/768/1024). Depths are set at 18/18/36 to match the number of parameters and FLOPs. The block structure remains the same (Fig. 4). We use the supervised training results from DeiT [73] for ViT-S/B and MAE [26] for ViT-L, as they employ improved training procedures over the original ViTs [20]. ConvNeXt models are trained with the same settings as before, but with longer warmup epochs. Results for ImageNet-1K at 2242 resolution are in Table 2. We observe ConvNeXt can perform generally on par with ViT, showing that our ConvNeXt block design is competitive when used in non-hierarchical models.

在這個消融中，我們研究了我們的ConvNeXt塊設計是否可以推廣到ViT式(ViT-style)的各向異性架構，這種架構沒有下采樣層，在所有深度都保持相同的特征分辨率（如14×14）。我們使用與ViT-S/B/L相同的特征尺寸（384/768/1024）構建各向異性的ConvNeXt-S/B/L。深度設置為18/18/36，以匹配參數和FLOPs的數量。塊狀結構保持不變（圖4）。

我們對ViT-S/B使用DeiT、的監督訓練結果，對ViT-L使用MAE[26]的監督訓練結果，因為它們采用了比原始ViTs[20]更好的訓練程序。ConvNeXt模型的訓練設置與之前相同，但有更長的預熱周期。表2列出了2242分辨率的ImageNet-1K的結果。我們觀察到ConvNeXt的表現基本與ViT持平，這表明我們的ConvNeXt塊設計在用于非層次模型時具有競爭力（non-hierarchical models）。

Table 2. Comparing isotropic ConvNeXt and ViT. Training memory is measured on V100 GPUs with 32 per-GPU batch size.
表2. 比較各向同性的ConvNeXt和ViT。訓練內存是在V100 GPU上測量的，每個GPU的批量大小為32。

4. Empirical Evaluation on Downstream Tasks —— 下游任務的實證評估

Object detection and segmentation on COCO

We finetune Mask R-CNN [27] and Cascade Mask R-CNN [9] on the COCO dataset with ConvNeXt backbones. Following Swin Transformer [45], we use multi-scale training, AdamW optimizer, and a 3× schedule. Further details and hyperparameter settings can be found in Appendix A.3.

我們在COCO數據集上用ConvNeXt骨干網絡對Mask R-CNN和Cascade Mask R-CNN進行微調。在Swin Transformer之后，我們使用了多尺度訓練、AdamW優化器和3×時間表。進一步的細節和超參數設置可以在附錄A.3中找到。

Table 3 shows object detection and instance segmentation results comparing Swin Transformer, ConvNeXt, and traditional ConvNet such as ResNeXt. Across different model complexities, ConvNeXt achieves on-par or better performance than Swin Transformer. When scaled up to bigger models (ConvNeXt-B/L/XL) pre-trained on ImageNet-22K, in many cases ConvNeXt is significantly better (e.g. +1.0 AP) than Swin Transformers in terms of box and mask AP.

表3顯示了Swin Transformer、ConvNeXt和ResNeXt等傳統ConvNet的物體檢測和實例分割結果的比較。

Table 3. COCO object detection and segmentation results using Mask-RCNN and Cascade Mask-RCNN. ? indicates that the model is pre-trained on ImageNet-22K. ImageNet-1K pre-trained Swin results are from their Github repository [3]. AP numbers of the ResNet-50 and X101 models are from [45]. We measure FPS on an A100 GPU. FLOPs are calculated with image size (1280, 800).
表3. 使用Mask-RCNN和Cascade Mask-RCNN進行COCO物體檢測和分割的結果。?表示該模型是在ImageNet-22K上預訓練的。ImageNet-1K的預訓練Swin結果來自其Github資源庫[3]。ResNet-50和X101模型的AP編號來自[45]。我們在A100 GPU上測量FPS。FLOPs是以圖像尺寸（1280，800）計算的。

在不同的模型復雜性中，ConvNeXt取得了與Swin Transformer相當或更好的性能。當擴大到在ImageNet-22K上預訓練的更大的模型（ConvNeXt-B/L/XL）時，在許多情況下，ConvNeXt在box 和mask AP方面明顯優于Swin Transformer（例如+1.0AP）。

Semantic segmentation on ADE20K

We also evaluate ConvNeXt backbones on the ADE20K semantic segmentation task with UperNet [85]. All model variants are trained for 160K iterations with a batch size of 16. Other experimental settings follow [6] (see Appendix A.3 for more details). In Table 4, we report validation mIoU with multi-scale testing. ConvNeXt models can achieve competitive performance across different model capacities, further validating the effectiveness of our architecture design.

我們還在ADE20K語義分割任務中評估了ConvNeXt骨架與UperNet的關系。所有的模型變體都訓練了16萬次迭代，批次大小為16。其他實驗設置遵循[6]（更多細節見附錄A.3）。在表4中，我們報告了多尺度測試的驗證mIoU。

Table 4. ADE20K validation results using UperNet [85]. ? indicates IN-22K pre-training. Swins’ results are from its GitHub repository [2]. Following Swin, we report mIoU results with multiscale testing. FLOPs are based on input sizes of (2048, 512) and (2560, 640) for IN-1K and IN-22K pre-trained models, respectively.
表4. 使用UperNet[85]的ADE20K驗證結果。?表示IN-22K預訓練。Swins的結果來自其GitHub倉庫[2]。繼Swin之后，我們報告了多尺度測試的mIoU結果。FLOPs是基于IN-1K和IN-22K預訓練模型的輸入尺寸（2048，512）和（2560，640）。

ConvNeXt模型可以在不同的模型容量下取得有競爭力的性能，進一步驗證了我們架構設計的有效性。

Remarks on model efficiency

Under similar FLOPs, models with depthwise convolutions are known to be slower and consume more memory than ConvNets with only dense convolutions. It is natural to ask whether the design of ConvNeXt will render it practically inefficient. As demonstrated throughout the paper, the inference throughputs of ConvNeXts are comparable to or exceed that of Swin Transformers. This is true for both classification and other tasks requiring higher-resolution inputs (see Table 1,3 for comparisons of throughput/FPS). Furthermore, we notice that training ConvNeXts requires less memory than training Swin Transformers. For example, training Cascade Mask-RCNN using ConvNeXt-B backbone consumes 17.4GB of peak memory with a per-GPU batch size of 2, while the reference number for Swin-B is 18.5GB. In comparison to vanilla ViT, both ConvNeXt and Swin Transformer exhibit a more favorable accuracy-FLOPs trade-off due to the local computations. It is worth noting that this improved efficiency is a result of the ConvNet inductive bias, and is not directly related to the self-attention mechanism in vision Transformers.

在類似的FLOPs下，已知具有深度卷積的模型比只有密集卷積的ConvNets更慢，消耗更多的內存。我們很自然地會問，ConvNeXt的設計是否會使其實際效率降低。正如本文所展示的那樣，ConvNeXt的推理吞吐量與Swin Transformers相當，甚至超過了Swin Transformers。這對于分類和其他需要高分辨率輸入的任務來說都是如此（吞吐量/FPS的比較見表1,3）。此外，我們注意到，訓練ConvNeXts需要的內存比訓練Swin Transformers少。例如，使用ConvNeXt-B骨干訓練Cascade Mask-RCNN，在每個GPU批次大小為2的情況下，消耗了17.4GB的峰值內存，而Swin-B的參考數字是18.5GB。與vanilla ViT相比，由于本地計算，ConvNeXt和Swin Transformer都表現出更有利的精度-FLOPs權衡。值得注意的是，這種效率的提高是ConvNet歸納偏置的結果，而與視覺Transformer中的自注意機制沒有直接關系。

5. Related Work

5.1 Hybrid models

In both the pre- and post-ViT eras, the hybrid model combining convolutions and self-attentions has been actively studied. Prior to ViT, the focus was on augmenting a ConvNet with self-attention/non-local modules [8, 55, 66, 79] to capture long-range dependencies. The original ViT [20] first studied a hybrid configuration, and a large body of follow-up works focused on reintroducing convolutional priors to ViT, either in an explicit [15, 16, 21, 82, 86, 88] or implicit [45] fashion.

在ViT之前和之后的時代，結合卷積和自留地的混合模型一直被積極研究。在ViT之前，重點是用自注意力/非本地模塊來增強ConvNet[8, 55, 66, 79]，以捕捉長距離的依賴關系。最初的ViT[20]首次研究了一種混合配置，大量的后續工作集中在將卷積先驗重新引入ViT，無論是以顯式[15, 16, 21, 82, 86, 88]還是隱式[45]方式。

5.2 Recent convolution-based approaches

Han et al. [25] show that local Transformer attention is equivalent to inhomogeneous dynamic depthwise conv. The MSA block in Swin is then replaced with a dynamic or regular depthwise convolution, achieving comparable performance to Swin. A concurrent work ConvMixer [4] demonstrates that, in small-scale settings, depthwise convolution can be used as a promising mixing strategy. ConvMixer uses a smaller patch size to achieve the best results, making the throughput much lower than other baselines. GFNet [56] adopts Fast Fourier Transform (FFT) for token mixing. FFT is also a form of convolution, but with a global kernel size and circular padding. Unlike many recent Transformer or ConvNet designs, one primary goal of our study is to provide an in-depth look at the process of modernizing a standard ResNet and achieving state-of-the-art performance.

Han等人[25]表明，局部Transformer注意力等同于不均勻的動態深度卷積，然后用動態或常規深度卷積取代Swin中的MSA塊，取得與Swin相當的性能。同時進行的一項工作ConvMixer[4]表明，在小范圍內，深度卷積可以作為一種有前途的混合策略。ConvMixer使用較小的補丁尺寸來達到最佳效果，使得吞吐量比其他基線低很多。GFNet[56]采用快速傅里葉變換（FFT）進行標記混合。FFT也是卷積的一種形式，但有一個全局內核大小和循環填充。與許多最近的Transformer或ConvNet設計不同，我們研究的一個主要目標是深入研究標準ResNet的現代化過程并實現最先進的性能。

6. Conclusions

In the 2020s, vision Transformers, particularly hierarchical ones such as Swin Transformers, began to overtake ConvNets as the favored choice for generic vision backbones. The widely held belief is that vision Transformers are more accurate, efficient, and scalable than ConvNets. We propose ConvNeXts, a pure ConvNet model that can compete favorably with state-of-the-art hierarchical vision Transformers across multiple computer vision benchmarks, while retaining the simplicity and efficiency of standard ConvNets. In some ways, our observations are surprising while our ConvNeXt model itself is not completely new — many design choices have all been examined separately over the last decade, but not collectively. We hope that the new results reported in this study will challenge several widely held views and prompt people to rethink the importance of convolution in computer vision.

在2020年代，視覺Transformer，特別是層次化的Transformer，如Swin Transformers，開始超越ConvNets，成為通用視覺骨干的首選。人們普遍認為，視覺Transformer比ConvNets更準確、更高效、更可擴展。我們提出了ConvNeXts，一個純ConvNet模型，它可以在多個計算機視覺基準中與最先進的分層視覺Transformer競爭，同時保留了標準ConvNets的簡單性和效率。在某些方面，我們的觀察結果令人驚訝，而我們的ConvNeXt模型本身并不是全新的——許多設計選擇都在過去十年中被單獨研究過，但沒有集體研究過。我們希望本研究報告的新結果將挑戰幾個廣泛持有的觀點，并促使人們重新思考計算機視覺中卷積的重要性。

Acknowledgments

We thank Kaiming He, Eric Mintun, Xingyi Zhou, Ross Girshick, and Yann LeCun for valuable discussions and feedback.

我們感謝何開明、Eric Mintun、周欣怡、Ross Girshick和Yann LeCun的寶貴討論和反饋。

Appendix —— 附錄

In this Appendix, we provide further experimental details (§A), robustness evaluation results (§B), more modernization experiment results (§C), and a detailed network specification (§D). We further benchmark model throughput on A100 GPUs (§E). Finally, we discuss the limitations (§F) and societal impact (§G) of our work.

在這個附錄中，我們提供了進一步的實驗細節（§A），魯棒性評估結果（§B），更多的現代化實驗結果（§C），以及詳細的網絡規范（§D）。我們進一步對A100 GPU上的模型吞吐量進行了基準測試（§E）。最后，我們討論了我們工作的局限性（§F）和社會影響（§G）。

A. Experimental Settings

A.1. ImageNet (Pre-)training

We provide ConvNeXts’ ImageNet-1K training and ImageNet-22K pre-training settings in Table 5. The settings are used for our main results in Table 1 (Section 3.2). All ConvNeXt variants use the same setting, except the stochastic depth rate is customized for model variants.

我們在表5中提供了ConvNeXts的ImageNet-1K訓練和ImageNet-22K預訓練設置。這些設置用于我們在表1（第3.2節）的主要結果。所有的ConvNeXt變體都使用相同的設置，只是隨機深度率是為模型變體定制的。

Table 5. ImageNet-1K/22K (pre-)training settings. Multiple stochastic depth rates (e.g., 0.1/0.4/0.5/0.5) are for each model (e.g., ConvNeXt-T/S/B/L) respectively.
表5. ImageNet-1K/22K（預）訓練設置。多個隨機深度率（如0.1/0.4/0.5/0.5）分別為每個模型（如ConvNeXt-T/S/B/L）。

For experiments in “modernizing a ConvNet” (Section 2), we also use Table 5’s setting for ImageNet-1K, except EMA is disabled, as we find using EMA severely hurts models with BatchNorm layers. For isotropic ConvNeXts (Section 3.3), the setting for ImageNet-1K in Table A is also adopted, but warmup is extended to 50 epochs, and layer scale is disabled for isotropic ConvNeXt-S/B. The stochastic depth rates are 0.1/0.2/0.5 for isotropic ConvNeXt-S/B/L.

在 "ConvNet現代化 "的實驗中（第2節），我們也使用了表5對ImageNet-1K的設置，只是EMA被禁用，因為我們發現使用EMA會嚴重傷害帶有BatchNorm層的模型。對于各向同性的ConvNeXts（第3.3節），我們也采用了表A中對ImageNet-1K的設置，但預熱時間延長到50個歷時，并且對于各向同性的ConvNeXt-S/B來說，層規模是禁用的。各向同性的ConvNeXt-S/B/L的隨機深度率為0.1/0.2/0.5。

A.2. ImageNet Fine-tuning

We list the settings for fine-tuning on ImageNet-1K in Table 6. The fine-tuning starts from the final model weights obtained in pre-training, without using the EMA weights, even if in pre-training EMA is used and EMA accuracy is reported. This is because we do not observe improvement if we fine-tune with the EMA weights (consistent with observations in [73]). The only exception is ConvNeXt-L pre-trained on ImageNet-1K, where the model accuracy is significantly lower than the EMA accuracy due to overfitting, and we select its best EMA model during pre-training as the starting point for fine-tuning.

我們在表6中列出了ImageNet-1K的微調設置。微調是從預訓練中得到的最終模型權重開始的，沒有使用EMA權重，即使在預訓練中使用了EMA，并且報告了EMA精度。這是因為如果使用EMA權重進行微調，我們并沒有觀察到改進（與[73]中的觀察一致）。唯一的例外是在ImageNet-1K上預訓練的ConvNeXt-L，由于過擬合，其模型精度明顯低于EMA精度，我們在預訓練中選擇其最佳EMA模型作為微調的起點。

In fine-tuning, we use layer-wise learning rate decay [6, 12] with every 3 consecutive blocks forming a group. When the model is fine-tuned at 384² resolution, we use a crop ratio of 1.0 (i.e., no cropping) during testing following [2, 74, 80], instead of 0.875 at 224².

在微調中，我們使用層間學習率衰減[6, 12]，每3個連續的塊形成一個組。當模型在384²分辨率下進行微調時，我們在測試過程中使用1.0的裁剪率（即不裁剪），而不是224²時的0.875。

A.3. Downstream Tasks

For ADE20K and COCO experiments, we follow the training settings used in BEiT [6] and Swin [45]. We also use MMDetection [10] and MMSegmentation [13] toolboxes. We use the final model weights (instead of EMA weights) from ImageNet pre-training as network initializations.

對于ADE20K和COCO的實驗，我們遵循BEiT[6]和Swin[45]中使用的訓練設置。我們還使用了MMDetection[10]和MMSegmentation[13]工具箱。我們使用ImageNet預訓練的最終模型權重（而不是EMA權重）作為網絡初始化。

We conduct a lightweight sweep for COCO experiments including learning rate {1e-4, 2e-4}, layer-wise learning rate decay [6] {0.7, 0.8, 0.9, 0.95}, and stochastic depth rate {0.3, 0.4, 0.5, 0.6, 0.7, 0.8}. We fine-tune the ImageNet-22K pre-trained Swin-B/L on COCO using the same sweep. We use the official code and pre-trained model weights [3].

我們對COCO實驗進行了輕量級掃描，包括學習率{ $\times 10^{-4}$ , $\times 10^{-4}$ }，層間學習率衰減[6] {0.7, 0.8, 0.9, 0.95}，以及隨機深度率{0.3, 0.4, 0.5, 0.6, 0.7, 0.8}。我們在COCO上使用同樣的掃頻對ImageNet-22K預訓練的Swin-B/L進行微調。我們使用官方代碼和預訓練的模型權重[3]。

The hyperparameters we sweep for ADE20K experiments include learning rate {8e-5, 1e-4}, layer-wise learning rate decay {0.8, 0.9}, and stochastic depth rate {0.3, 0.4, 0.5}. We report validation mIoU results using multi-scale testing. Additional single-scale testing results are in Table 7.

我們為ADE20K實驗掃除的超參數包括學習率{ $\times 10^{-5}$ , $\times 10^{-4}$ }，層間學習率衰減{0.8, 0.9}，以及隨機深度率{0.3, 0.4, 0.5}。我們報告了使用多尺度測試的驗證性mIoU結果。其他單尺度測試結果見表7。

B. Robustness Evaluation

Additional robustness evaluation results for ConvNeXt models are presented in Table 8. We directly test our ImageNet-1K trained/fine-tuned classification models on several robustness benchmark datasets such as ImageNet-A [33], ImageNet-R [30], ImageNet-Sketch [78] and ImageNetC/ $Cˉ\bar{\mathrm{C}}$ [31, 48] datasets. We report mean corruption error (mCE) for ImageNet-C, corruption error for ImageNet- $Cˉ\bar{\mathrm{C}}$ , and top-1 Accuracy for all other datasets.

表8中列出了ConvNeXt模型的其他魯棒性評估結果。我們直接在幾個魯棒性基準數據集上測試我們的ImageNet-1K訓練/微調分類模型，如ImageNet-A [33], ImageNet-R [30], ImageNet-Sketch [78] 和ImageNetC/ $Cˉ\bar{\mathrm{C}}$ [31, 48] 數據集。我們報告了ImageNet-C的平均腐蝕誤差（mCE），ImageNet- $Cˉ\bar{\mathrm{C}}$ 的腐蝕誤差，以及所有其他數據集的top-1準確率。

Table 8. Robustness evaluation of ConvNeXt. We do not make use of any specialized modules or additional fine-tuning procedures.
表8. ConvNeXt的魯棒性評估。我們沒有使用任何專門的模塊或額外的微調程序。

ConvNeXt (in particular the large-scale model variants) exhibits promising robustness behaviors, outperforming state-of-the-art robust transformer models [47] on several benchmarks. With extra ImageNet-22K data, ConvNeXtXL demonstrates strong domain generalization capabilities (e.g. achieving 69.3%/68.2%/55.0% accuracy on ImageNetA/R/Sketch benchmarks, respectively). We note that these robustness evaluation results were acquired without using any specialized modules or additional fine-tuning procedures.

ConvNeXt（尤其是大規模模型的變體）表現出了很好的魯棒性行為，在一些基準測試上超過了最先進的魯棒性Transformer模型[47]。利用額外的ImageNet-22K數據，ConvNeXt XL展示了強大的領域泛化能力（例如，在ImageNetA/R/Sketch基準上分別達到69.3%/68.2%/55.0%的精度）。我們注意到，這些魯棒性評估結果是在沒有使用任何專門模塊或額外微調程序的情況下獲得的。

C. Modernizing ResNets: detailed results

Here we provide detailed tabulated results for the modernization experiments, at both ResNet-50 / Swin-T and ResNet-200 / Swin-B regimes. The ImageNet-1K top-1 accuracies and FLOPs for each step are shown in Table 10 and 11. ResNet-50 regime experiments are run with 3 random seeds.

這里我們提供了在ResNet-50 / Swin-T和ResNet-200 / Swin-B兩個制度下的現代化實驗的詳細表格結果。表10和11顯示了ImageNet-1K每一步的最高準確率和FLOPs。ResNet-50制度的實驗是用3個隨機種子運行的。

For ResNet-200, the initial number of blocks at each stage is (3, 24, 36, 3). We change it to Swin-B’s (3, 3, 27, 3) at the step of changing stage ratio. This drastically reduces the FLOPs, so at the same time, we also increase the width from 64 to 84 to keep the FLOPs at a similar level. After the step of adopting depthwise convolutions, we further increase the width to 128 (same as Swin-B’s) as a separate step.

對于ResNet-200，每個階段的初始塊數是（3, 24, 36, 3）。在改變階段比例的步驟中，我們將其改為Swin-B的（3, 3, 27, 3）。這大大減少了FLOPs，所以同時我們也將寬度從64增加到84，以保持FLOPs在一個類似的水平。在采用深度卷積的步驟后，我們進一步將寬度增加到128（與Swin-B的相同），作為一個單獨的步驟。

The observations on the ResNet-200 regime are mostly consistent with those on ResNet-50 as described in the main paper. One interesting difference is that inverting dimensions brings a larger improvement at ResNet-200 regime than at ResNet-50 regime (+0.79% vs. +0.14%). The performance gained by increasing kernel size also seems to saturate at kernel size 5 instead of 7. Using fewer normalization layers also has a bigger gain compared with the ResNet-50 regime (+0.46% vs. +0.14%).

對ResNet-200系統的觀察與主論文中描述的ResNet-50系統的觀察基本一致。一個有趣的區別是，與ResNet-50系統相比，倒置尺寸帶來了更大的改進（+0.79% vs. +0.14%）。與ResNet-50系統相比，使用較少的歸一化層也有更大的收益（+0.46% vs. +0.14%）。

D. Detailed Architectures

We present a detailed architecture comparison between ResNet-50, ConvNeXt-T and Swin-T in Table 9. For differently sized ConvNeXts, only the number of blocks and the number of channels at each stage differ from ConvNeXt-T (see Section 3 for details). ConvNeXts enjoy the simplicity of standard ConvNets, but compete favorably with Swin Transformers in visual recognition.

我們在表9中列出了ResNet-50、ConvNeXt-T和Swin-T之間的詳細結構比較。對于不同大小的ConvNeXts，只有每個階段的塊數和通道數與ConvNeXt-T不同（詳見第三節）。ConvNeXts享有標準ConvNets的簡單性，但在視覺識別方面與Swin Transformers的競爭很有利。

E. Benchmarking on A100 GPUs

Following Swin Transformer [45], the ImageNet models’ inference throughputs in Table 1 are benchmarked using a V100 GPU, where ConvNeXt is slightly faster in inference than Swin Transformer with a similar number of parameters. We now benchmark them on the more advanced A100 GPUs, which support the TensorFloat32 (TF32) tensor cores. We employ PyTorch [50] version 1.10 to use the latest “Channel Last” memory layout [22] for further speedup.

按照Swin Transformer[45]的做法，表1中ImageNet模型的推理吞吐量是使用V100 GPU進行基準測試的，在參數數量相似的情況下，ConvNeXt的推理速度略高于Swin Transformer。現在我們在更先進的A100 GPU上對它們進行基準測試，它支持TensorFloat32（TF32）張量核心。我們采用PyTorch[50]1.10版本，使用最新的 "Channel Last "內存布局[22]，以進一步提高速度。

We present the results in Table 12. Swin Transformers and ConvNeXts both achieve faster inference throughput than V100 GPUs, but ConvNeXts’ advantage is now significantly greater, sometimes up to 49% faster. This preliminary study shows promising signals that ConvNeXt, employed with standard ConvNet modules and simple in design, could be practically more efficient models on modern hardwares.

我們在表12中列出了結果。Swin Transformers和ConvNeXts都取得了比V100 GPU更快的推理吞吐量，但ConvNeXts的優勢現在明顯更大，有時可以快到49%。這項初步研究顯示了有希望的信號，即ConvNeXt，采用標準的ConvNet模塊，設計簡單，實際上可以在現代硬軟件上成為更有效的模型。

Table 12. Inference throughput comparisons on an A100 GPU. Using TF32 data format and “channel last” memory layout, ConvNeXt enjoys up to ～49% higher throughput compared with a Swin Transformer with similar FLOPs.
表12. A100 GPU上的推理吞吐量比較。使用TF32數據格式和 "通道最后(channel last) "內存布局，ConvNeXt與具有類似FLOPs的Swin Transformer相比，享有高達49%的吞吐量。

F. Limitations

We demonstrate ConvNeXt, a pure ConvNet model, can perform as good as a hierarchical vision Transformer on image classification, object detection, instance and semantic segmentation tasks. While our goal is to offer a broad range of evaluation tasks, we recognize computer vision applications are even more diverse. ConvNeXt may be more suited for certain tasks, while Transformers may be more flexible for others. A case in point is multi-modal learning, in which a cross-attention module may be preferable for modeling feature interactions across many modalities. Additionally, Transformers may be more flexible when used for tasks requiring discretized, sparse, or structured outputs. We believe the architecture choice should meet the needs of the task at hand while striving for simplicity.

我們證明了ConvNeXt，一個純粹的ConvNet模型，在圖像分類、物體檢測、實例和語義分割等任務上的表現不亞于層次化的視覺變換器。雖然我們的目標是提供廣泛的評估任務，但我們認識到計算機視覺的應用甚至更加多樣化。ConvNeXt可能更適合某些任務，而Transformer可能對其他任務更靈活。一個典型的例子是多模態學習，在這種情況下，交叉注意力模塊可能更適合于為許多模態之間的特征互動建模。此外，當用于需要離散的(discretized)、稀疏的(sparse)或結構化(structured )輸出的任務時，Transformer可能更靈活。我們認為，架構的選擇應該滿足手頭任務的需要，同時爭取做到簡單。

G. Societal Impact

In the 2020s, research on visual representation learning began to place enormous demands on computing resources. While larger models and datasets improve performance across the board, they also introduce a slew of challenges. ViT, Swin, and ConvNeXt all perform best with their huge model variants. Investigating those model designs inevitably results in an increase in carbon emissions. One important direction, and a motivation for our paper, is to strive for simplicity — with more sophisticated modules, the network’s design space expands enormously, obscuring critical components that contribute to the performance difference. Additionally, large models and datasets present issues in terms of model robustness and fairness. Further investigation on the robustness behavior of ConvNeXt vs. Transformer will be an interesting research direction. In terms of data, our findings indicate that ConvNeXt models benefit from pre-training on large-scale datasets. While our method makes use of the publicly available ImageNet-22K dataset, individuals may wish to acquire their own data for pre-training. A more circumspect and responsible approach to data selection is required to avoid potential concerns with data biases.

在2020年代，關于視覺表征學習的研究開始對計算資源提出了巨大的要求。雖然更大的模型和數據集全面提高了性能，但也帶來了一系列的挑戰。ViT、Swin和ConvNeXt都在其巨大的模型變體中表現最好。研究這些模型設計不可避免地會導致碳排放的增加。一個重要的方向，也是我們論文的動機，就是力求簡單——隨著更復雜的模塊，網絡的設計空間會極大地擴展，掩蓋了造成性能差異的關鍵部件。此外，大型模型和數據集在模型魯棒性和公平性方面存在問題。對ConvNeXt與Transformer的魯棒性行為的進一步調查將是一個有趣的研究方向。在數據方面，我們的發現表明ConvNeXt模型得益于大規模數據集的預訓練。雖然我們的方法利用了公開的ImageNet-22K數據集，但個人可能希望獲得自己的數據進行預訓練。為避免潛在的數據偏差問題，需要采取更加謹慎和負責任的方法來選擇數據。

References

[1] PyTorch Vision Models. https://pytorch.org/vision/stable/models.html. Accessed: 2021-10-01.
[2] GitHub repository: Swin transformer. https://github.com/microsoft/Swin-Transformer, 2021.
[3] GitHub repository: Swin transformer for object detection.https://github.com/SwinTransformer/Swin-Transformer-Object-Detection, 2021.
[4] Anonymous. Patches are all you need? Openreview, 2021.
[5] Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E Hinton. Layer normalization. arXiv:1607.06450, 2016.
[6] Hangbo Bao, Li Dong, and Furu Wei. BEiT: BERT pre-training of image transformers. arXiv:2106.08254, 2021.
[7] Irwan Bello, William Fedus, Xianzhi Du, Ekin Dogus Cubuk, Aravind Srinivas, Tsung-Yi Lin, Jonathon Shlens, and Barret Zoph. Revisiting resnets: Improved training and scaling strategies. NeurIPS, 2021.
[8] Irwan Bello, Barret Zoph, Ashish Vaswani, Jonathon Shlens, and Quoc V Le. Attention augmented convolutional networks. In ICCV, 2019.
[9] Zhaowei Cai and Nuno Vasconcelos. Cascade R-CNN: Delving into high quality object detection. In CVPR, 2018.
[10] Kai Chen, Jiaqi Wang, Jiangmiao Pang, Yuhang Cao, Yu Xiong, Xiaoxiao Li, Shuyang Sun, Wansen Feng, Ziwei Liu, Jiarui Xu, Zheng Zhang, Dazhi Cheng, Chenchen Zhu, Tianheng Cheng, Qijie Zhao, Buyu Li, Xin Lu, Rui Zhu, Yue Wu, Jifeng Dai, Jingdong Wang, Jianping Shi, Wanli Ouyang, Chen Change Loy, and Dahua Lin. MMDetection: Open mmlab detection toolbox and benchmark. arXiv:1906.07155, 2019.
[11] Fran?ois Chollet. Xception: Deep learning with depthwise separable convolutions. In CVPR, 2017.
[12] Kevin Clark, Minh-Thang Luong, Quoc V Le, and Christopher D Manning. ELECTRA: Pre-training text encoders as discriminators rather than generators. In ICLR, 2020.
[13] MMSegmentation contributors. MMSegmentation: Openmmlab semantic segmentation toolbox and benchmark. https://github.com/open-mmlab/mmsegmentation, 2020.
[14] Ekin D Cubuk, Barret Zoph, Jonathon Shlens, and Quoc V Le. Randaugment: Practical automated data augmentation with a reduced search space. In CVPR Workshops, 2020.
[15] Zihang Dai, Hanxiao Liu, Quoc V Le, and Mingxing Tan. Coatnet: Marrying convolution and attention for all data sizes. NeurIPS, 2021.
[16] Stéphane d’Ascoli, Hugo Touvron, Matthew Leavitt, Ari Morcos, Giulio Biroli, and Levent Sagun. ConViT: Improving vision transformers with soft convolutional inductive biases. ICML, 2021.
[17] Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. ImageNet: A large-scale hierarchical image database. In CVPR, 2009.
[18] Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT: Pre-training of deep bidirectional transformers for language understanding. In NAACL, 2019.
[19] Piotr Dollár, Serge Belongie, and Pietro Perona. The fastest pedestrian detector in the west. In BMVC, 2010.
[20] Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. In ICLR, 2021.
[21] Haoqi Fan, Bo Xiong, Karttikeya Mangalam, Yanghao Li, Zhicheng Yan, Jitendra Malik, and Christoph Feichtenhofer. Multiscale vision transformers. ICCV, 2021.
[22] Vitaly Fedyunin. Tutorial: Channel last memory format in PyTorch. https://pytorch.org/tutorials/intermediate/memory_format_tutorial.html, 2021. Accessed: 2021-10-01.
[23] Ross Girshick. Fast R-CNN. In ICCV, 2015.
[24] Ross Girshick, Jeff Donahue, Trevor Darrell, and Jitendra Malik. Rich feature hierarchies for accurate object detection and semantic segmentation. In CVPR, 2014.
[25] Qi Han, Zejia Fan, Qi Dai, Lei Sun, Ming-Ming Cheng, Jiaying Liu, and Jingdong Wang. Demystifying local vision transformer: Sparse connectivity, weight sharing, and dynamic weight. arXiv:2106.04263, 2021.
[26] Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. Masked autoencoders are scalable vision learners. arXiv:2111.06377, 2021.
[27] Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick. Mask R-CNN. In ICCV, 2017.
[28] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR, 2016.
[29] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Identity mappings in deep residual networks. In ECCV, 2016.
[30] Dan Hendrycks, Steven Basart, Norman Mu, Saurav Kadavath, Frank Wang, Evan Dorundo, Rahul Desai, Tyler Zhu, Samyak Parajuli, Mike Guo, et al. The many faces of robustness: A critical analysis of out-of-distribution generalization. In ICCV, 2021.
[31] Dan Hendrycks and Thomas Dietterich. Benchmarking neural network robustness to common corruptions and perturbations. In ICLR, 2018.
[32] Dan Hendrycks and Kevin Gimpel. Gaussian error linear units (gelus). arXiv:1606.08415, 2016.
[33] Dan Hendrycks, Kevin Zhao, Steven Basart, Jacob Steinhardt, and Dawn Song. Natural adversarial examples. In CVPR, 2021.
[34] Andrew G Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861, 2017.
[35] Jie Hu, Li Shen, and Gang Sun. Squeeze-and-excitation networks. In CVPR, 2018.
[36] Gao Huang, Zhuang Liu, Laurens van der Maaten, and Kilian Q Weinberger. Densely connected convolutional networks. In CVPR, 2017.
[37] Gao Huang, Yu Sun, Zhuang Liu, Daniel Sedra, and Kilian Q Weinberger. Deep networks with stochastic depth. In ECCV, 2016.
[38] Sergey Ioffe. Batch renormalization: Towards reducing minibatch dependence in batch-normalized models. In NeurIPS, 2017.
[39] Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Joan Puigcerver, Jessica Yung, Sylvain Gelly, and Neil Houlsby. Big Transfer (BiT): General visual representation learning. In ECCV, 2020.
[40] Alex Krizhevsky, Ilya Sutskever, and Geoff Hinton. Imagenet classification with deep convolutional neural networks. In NeurIPS, 2012.
[41] Andrew Lavin and Scott Gray. Fast algorithms for convolutional neural networks. In CVPR, 2016.
[42] Yann LeCun, Bernhard Boser, John S Denker, Donnie Henderson, Richard E Howard, Wayne Hubbard, and Lawrence D Jackel. Backpropagation applied to handwritten zip code recognition. Neural computation, 1989.
[43] Yann LeCun, Léon Bottou, Yoshua Bengio, Patrick Haffner, et al. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 1998.
[44] Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. Microsoft COCO: Common objects in context. In ECCV. 2014.
[45] Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. Swin transformer: Hierarchical vision transformer using shifted windows. 2021.
[46] Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. In ICLR, 2019.
[47] Xiaofeng Mao, Gege Qi, Yuefeng Chen, Xiaodan Li, Ranjie Duan, Shaokai Ye, Yuan He, and Hui Xue. Towards robust vision transformer. arXiv preprint arXiv:2105.07926, 2021.
[48] Eric Mintun, Alexander Kirillov, and Saining Xie. On interaction between augmentations and corruptions in natural corruption robustness. NeurIPS, 2021.
[49] Vinod Nair and Geoffrey E Hinton. Rectified linear units improve restricted boltzmann machines. In ICML, 2010.
[50] Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. PyTorch: An imperative style, high-performance deep learning library. In NeurIPS, 2019.
[51] Boris T Polyak and Anatoli B Juditsky. Acceleration of stochastic approximation by averaging. SIAM Journal on Control and Optimization, 1992.
[52] Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. Language models are unsupervised multitask learners. 2019.
[53] Ilija Radosavovic, Justin Johnson, Saining Xie, Wan-Yen Lo, and Piotr Dollár. On network design spaces for visual recognition. In ICCV, 2019.
[54] Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, and Piotr Dollár. Designing network design spaces. In CVPR, 2020.
[55] Prajit Ramachandran, Niki Parmar, Ashish Vaswani, Irwan Bello, Anselm Levskaya, and Jonathon Shlens. Stand-alone self-attention in vision models. NeurIPS, 2019.
[56] Yongming Rao, Wenliang Zhao, Zheng Zhu, Jiwen Lu, and Jie Zhou. Global filter networks for image classification. NeurIPS, 2021.
[57] Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster R-CNN: Towards real-time object detection with region proposal networks. In NeurIPS, 2015.
[58] Henry A Rowley, Shumeet Baluja, and Takeo Kanade. Neural network-based face detection. TPAMI, 1998.
[59] Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg, and Li Fei-Fei. ImageNet Large Scale Visual Recognition Challenge. IJCV, 2015.
[60] Tim Salimans and Diederik P Kingma. Weight normalization: A simple reparameterization to accelerate training of deep neural networks. In NeurIPS, 2016.
[61] Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. Mobilenetv2: Inverted residuals and linear bottlenecks. In CVPR, 2018.
[62] Pierre Sermanet, David Eigen, Xiang Zhang, Michael Mathieu, Rob Fergus, and Yann LeCun. Overfeat: Integrated recognition, localization and detection using convolutional networks. In ICLR, 2014.
[63] Pierre Sermanet, Koray Kavukcuoglu, Soumith Chintala, and Yann LeCun. Pedestrian detection with unsupervised multistage feature learning. In CVPR, 2013.
[64] Karen Simonyan and Andrew Zisserman. Two-stream convolutional networks for action recognition in videos. In NeurIPS, 2014.
[65] Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. In ICLR, 2015.
[66] Aravind Srinivas, Tsung-Yi Lin, Niki Parmar, Jonathon Shlens, Pieter Abbeel, and Ashish Vaswani. Bottleneck transformers for visual recognition. In CVPR, 2021.
[67] Andreas Steiner, Alexander Kolesnikov, Xiaohua Zhai, Ross Wightman, Jakob Uszkoreit, and Lucas Beyer. How to train your vit? data, augmentation, and regularization in vision transformers. arXiv preprint arXiv:2106.10270, 2021.
[68] Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. Going deeper with convolutions. In CVPR, 2015.
[69] Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, and Zbigniew Wojna. Rethinking the inception architecture for computer vision. In CVPR, 2016.
[70] Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, Mark Sandler, Andrew Howard, and Quoc V Le. Mnasnet: Platform-aware neural architecture search for mobile. In CVPR, 2019.
[71] Mingxing Tan and Quoc Le. Efficientnet: Rethinking model scaling for convolutional neural networks. In ICML, 2019.
[72] Mingxing Tan and Quoc Le. Efficientnetv2: Smaller models and faster training. In ICML, 2021.
[73] Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, and Hervé Jégou. Training data-efficient image transformers & distillation through attention. arXiv:2012.12877, 2020.
[74] Hugo Touvron, Matthieu Cord, Alexandre Sablayrolles, Gabriel Synnaeve, and Hervé Jégou. Going deeper with image transformers. ICCV, 2021.
[75] Dmitry Ulyanov, Andrea Vedaldi, and Victor Lempitsky. Instance normalization: The missing ingredient for fast stylization. arXiv:1607.08022, 2016.
[76] Régis Vaillant, Christophe Monrocq, and Yann Le Cun. Original approach for the localisation of objects in images. Vision, Image and Signal Processing, 1994.
[77] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In NeurIPS, 2017.
[78] Haohan Wang, Songwei Ge, Eric P Xing, and Zachary C Lipton. Learning robust global representations by penalizing local predictive power. NeurIPS, 2019.
[79] Xiaolong Wang, Ross Girshick, Abhinav Gupta, and Kaiming He. Non-local neural networks. In CVPR, 2018.
[80] Ross Wightman. GitHub repository: Pytorch image models. https://github.com/rwightman/pytorchimage-models, 2019.
[81] Ross Wightman, Hugo Touvron, and Hervé Jégou. Resnet strikes back: An improved training procedure in timm. arXiv:2110.00476, 2021.
[82] Haiping Wu, Bin Xiao, Noel Codella, Mengchen Liu, Xiyang Dai, Lu Yuan, and Lei Zhang. Cvt: Introducing convolutions to vision transformers. ICCV, 2021.
[83] Yuxin Wu and Kaiming He. Group normalization. In ECCV, 2018.
[84] Yuxin Wu and Justin Johnson. Rethinking “batch” in batchnorm. arXiv:2105.07576, 2021.
[85] Tete Xiao, Yingcheng Liu, Bolei Zhou, Yuning Jiang, and Jian Sun. Unified perceptual parsing for scene understanding. In ECCV, 2018.
[86] Tete Xiao, Mannat Singh, Eric Mintun, Trevor Darrell, Piotr Dollár, and Ross Girshick. Early convolutions help transformers see better. In NeurIPS, 2021.
[87] Saining Xie, Ross Girshick, Piotr Dollár, Zhuowen Tu, and Kaiming He. Aggregated residual transformations for deep neural networks. In CVPR, 2017.
[88] Weijian Xu, Yifan Xu, Tyler Chang, and Zhuowen Tu. Coscale conv-attentional image transformers. ICCV, 2021.
[89] Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk
Chun, Junsuk Choe, and Youngjoon Yoo. Cutmix: Regularization strategy to train strong classifiers with localizable features. In ICCV, 2019.
[90] Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. mixup: Beyond empirical risk minimization. In ICLR, 2018.
[91] Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang. Random erasing data augmentation. In AAAI, 2020.
[92] Bolei Zhou, Hang Zhao, Xavier Puig, Tete Xiao, Sanja Fidler, Adela Barriuso, and Antonio Torralba. Semantic understanding of scenes through the ADE20K dataset. IJCV, 2019

總結

以上是生活随笔為你收集整理的A ConvNet for the 2020s的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。

ConvNet

上一篇： java中使用MD5验证文件的完整性
下一篇： oracle 数据库存储过程编译报错PL