DL之InceptionV2/V3：InceptionV2 & InceptionV3算法的簡介(論文介紹)、架構詳解、案例應用等配圖集合之詳細攻略

?

目錄

InceptionV2 & InceptionV3算法的簡介(論文介紹)

InceptionV2 & InceptionV3算法的架構詳解

1、卷積分解

2、Inception模塊

3、Inception v2 & v3網絡模塊

4、對Auxiliary Classifier(輔助分類器)的考慮

5、標簽平滑的模型正則化

6、Inception v3

InceptionV2 & InceptionV3算法的案例應用

---

?

?

相關文章
DL之GoogleNet：GoogleNet(InceptionV1)算法的簡介(論文介紹)、架構詳解、案例應用等配圖集合之詳細攻略
DL之BN-Inception：BN-Inception算法的簡介(論文介紹)、架構詳解、案例應用等配圖集合之詳細攻略
DL之InceptionV2/V3：InceptionV2 & InceptionV3算法的簡介(論文介紹)、架構詳解、案例應用等配圖集合之詳細攻略
DL之InceptionV2/V3：InceptionV2 & InceptionV3算法的架構詳解???????
DL之InceptionV4/ResNet：InceptionV4/Inception-ResNet算法的簡介(論文介紹)、架構詳解、案例應用等配圖集合之詳細攻略

InceptionV2 & InceptionV3算法的簡介(論文介紹)

? ? ? ?InceptionV2 & InceptionV3是谷歌研究人員，在InceptionV1和BN-Inception網絡模型基礎上進行改進的。

摘要
? ? ? Convolutional networks are at the core of most stateof-the-art ?computer vision solutions for a wide variety of ?tasks. Since 2014 very deep convolutional networks started ?to become mainstream, yielding substantial gains in various ?benchmarks. Although increased model size and computational ?cost tend to translate to immediate quality gains ?for most tasks (as long as enough labeled data is provided ?for training), computational efficiency and low parameter ?count are still enabling factors for various use cases such as ?mobile vision and big-data scenarios. Here we are exploring ?ways to scale up networks in ways that aim at utilizing ?the added computation as efficiently as possible by suitably ?factorized convolutions and aggressive regularization. We ?benchmark our methods on the ILSVRC 2012 classification ?challenge validation set demonstrate substantial gains over ?the state of the art: 21.2% top-1 and 5.6% top-5 error for ?single frame evaluation using a network with a computational ?cost of 5 billion multiply-adds per inference and with ?using less than 25 million parameters. With an ensemble of ?4 models and multi-crop evaluation, we report 3.5% top-5 ?error and 17.3% top-1 error.
? ? ? 卷積網絡是最先進的計算機視覺解決方案的核心，可用于各種各樣的任務。自2014年以來，非常深的卷積網絡開始成為主流，在各種基準中產生了實質性的收益。盡管增加的模型大小和計算成本往往會轉化為大多數任務的即時質量收益（只要為訓練提供足夠的標記數據），計算效率和低參數計數仍然是各種用例（如移動視覺和大數據場景。在這里，我們正在探索擴大網絡的方法，旨在通過適當的因子分解卷積和積極的正則化，盡可能有效地利用增加的計算。我們在ILSVRC 2012分類挑戰驗證集上對我們的方法進行了基準測試，結果表明，與最新技術相比，單幀評估的21.2%Top-1和5.6%Top-5錯誤顯著增加，使用的網絡計算成本為每次推理增加50億，并且使用少于2500萬個參數。綜合4個模型和多作物評估，我們報告了3.5% top-5 錯誤 and 17.3% top-1 錯誤。
結論
? ? ? ? We have provided several design principles to scale up ?convolutional networks and studied them in the context of ?the Inception architecture. This guidance can lead to high ?performance vision networks that have a relatively modest ?computation cost compared to simpler, more monolithic ?architectures. Our highest quality version of Inception-v3 ?reaches 21.2%, top-1 and 5.6% top-5 error for single crop ?evaluation on the ILSVR 2012 classification, setting a new ?state of the art. This is achieved with relatively modest ?(2.5×) increase in computational cost compared to the network ?described in Ioffe et al [7]. Still our solution uses ?much less computation than the best published results based ?on denser networks: our model outperforms the results of ?He et al [6] – cutting the top-5 (top-1) error by 25% (14%) ?relative, respectively – while being six times cheaper computationally ?and using at least five times less parameters ?(estimated). Our ensemble of four Inception-v3 models ?reaches 3.5% with multi-crop evaluation reaches 3.5% top5 ?error which represents an over 25% reduction to the best ?published results and is almost half of the error of ILSVRC ?2014 winining GoogLeNet ensemble. ?
? ? ? ? 我們提供了幾個擴展卷積網絡的設計原則，并在初始體系結構的上下文中對它們進行了研究。與更簡單、更單一的體系結構相比，這種指導可以導致具有相對較低計算成本的高性能視覺網絡。在ILSVR 2012分類的單作物評估中，我們最高質量版本的Inception-v3達到21.2%, top-1 and 5.6% top-5 錯誤，創造了新的技術水平。與Ioffe等人所述的網絡相比，這是通過相對適度（2.5倍）的計算成本增加實現的。盡管如此，我們的解決方案使用的計算量比基于更密集網絡的最佳公布結果要少得多：我們的模型比He等人的計算結果要好得多——分別將前5（前1）個錯誤相對減少了 top-5 (top-1) 錯誤 by 25% (14%)，同時計算成本低了6倍，并且至少使用了參數（估計）減少5倍。我們的四個初始-v3模型的集合達到3.5%，多作物評估達到3.5% top5的錯誤，這意味著最佳公布結果減少了25%以上，幾乎是ILSVRC 2014冠軍 GoogLeNet 集合誤差的一半。
? ? ? ? We have also demonstrated that high quality results can ?be reached with receptive field resolution as low as 79×79. ?This might prove to be helpful in systems for detecting relatively ?small objects. We have studied how factorizing convolutions ?and aggressive dimension reductions inside neural ?network can result in networks with relatively low computational ?cost while maintaining high quality. The combination ?of lower parameter count and additional regularization with ?batch-normalized auxiliary classifiers and label-smoothing ?allows for training high quality networks on relatively modest ?sized training sets.
? ? ? ? 我們還證明了接收場分辨率低至79×79可以獲得高質量的結果。這可能被證明有助于系統檢測相對較小的物體。我們研究了在保持高質量的同時，神經網絡中的因子分解卷積和積極的降維是如何產生計算成本相對較低的網絡的。將較低的參數計數和額外的正則化與批標準化輔助分類器和標簽平滑相結合，可以在相對較小的訓練集上訓練高質量的網絡。

?

論文
Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, Zbigniew Wojna.
Rethinking the Inception Architecture for Computer Vision
https://arxiv.org/abs/1512.00567

?

InceptionV2 & InceptionV3算法的架構詳解

DL之InceptionV2/V3：InceptionV2 & InceptionV3算法的架構詳解???????

?

?

?

InceptionV2 & InceptionV3算法的案例應用

TF之DD：實現輸出Inception模型內的某個卷積層或者所有卷積層的形狀
TF之DD：利用Inception模型+GD算法生成原始的Deep Dream圖片
TF之DD：利用Inception模型+GD算法生成更大尺寸的Deep Dream精美圖片
TF之DD：利用Inception模型+GD算法生成更高質量的Deep Dream高質量圖片
TF之DD：利用Inception模型+GD算法——五個架構設計思路
TF之DD：利用Inception模型+GD算法生成帶背景的大尺寸、高質量的Deep Dream圖片

?

?

?

?

總結

以上是生活随笔為你收集整理的DL之InceptionV2/V3：InceptionV2 InceptionV3算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略的全部內容，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。