The method or strategy used by SVM for multi-classification/使用SVM进行多分类时所用的方法或策略

[moombeam1]

When replicating your code, I didn't understand the multi-classification method you used when using SVM in UrbanSound8K classification. Do you use one-versus-one or one-versus-rest strategy, or something else. 在复现您的代码时，我对您在UrbanSound8K 分类中使用SVM进行分类时所用的多分类方法没有看明白，请问您使用的是一对一策略还是一对多策略，亦或是其他的方法。

[ppphhhleo]

Hi Moonbeam1,

Thanks for the feedback, as I haven't updated the repo since the repo was created, may I ask if you have replicated the codes successfully? I have reviewed the codes again, and the UrbanSound8K SVM classification in ClassifyAudio.ipynb was implemented by Sklearn repo, specifically sklearn.svm.SVC[2]. The implementation of sklearn.svm.SVC is based on libsvm[3], and the multiclass support is handled according to a one-vs-one[1][2] scheme. You may find more info in links related to sklearn.svm.SVC. If you find the repo helpful, feel free to star it, thanks :)

您好，感谢您的反馈，该代码仓库在两年前创建后没有更新，请问您是否成功复现了这个仓库代码？ 我再次查看了ClassifyAudio.ipynb中的代码，其中UrbanSound8K SVM 分类是由 Sklearn repo，具体是 sklearn.svm.SVC[2]实现的。 sklearn.svm.SVC 是基于 libsvm[3]实现的，多分类支持是按照 one-vs-one[1][2]方案处理的。 您可以在 sklearn.svm.SVC 相关链接中找到更多信息。 如果这代码仓库有帮到你，欢迎Star，感谢 :)

[1] Sklearn多分类：https://www.cnblogs.com/ruidongwu/p/11869948.html [2] Sklearn.svm.SVC: https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html [3] Libsvm: https://www.csie.ntu.edu.tw/~cjlin/libsvm/

Warmest, Leo Pan.

On Sun, Apr 14, 2024 at 5:59 PM moombeam1 @.***> wrote:

When replicating your code, I didn't understand the multi-classification method you used when using SVM in UrbanSound8K classification. Do you use one-versus-one or one-versus-rest strategy, or something else. 在复现您的代码时，我对您在UrbanSound8K 分类中使用SVM进行分类时所用的多分类方法没有看明白，请问您使用的是一对一策略还是一对多策略，亦或是其他的方法。

— Reply to this email directly, view it on GitHub https://github.com/ppphhhleo/Audio_practice/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO5RI2BKI4OQD3L4UFRW4MLY5JHQVAVCNFSM6AAAAABGGDAIJCVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI2DEMBXGY2DIMY . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[moombeam1]

您好 Moonbeam1，感谢您的反馈，因为自从创建仓库以来我还没有更新过仓库，请问您是否已成功复制代码？我再次查看了代码，ClassifyAudio.ipynb 中的 UrbanSound8K SVM 分类是由 Sklearn 存储库实现的，具体为 sklearn.svm.SVC[2]。sklearn.svm.SVC 的实现基于 libsvm[3]，并且根据 one-vs-one[1][2] 方案处理多类支持。您可以在与 sklearn.svm.SVC 相关的链接中找到更多信息。如果您发现该仓库有帮助，请随时为它加注星标，谢谢:)您好，感谢您的反馈，该代码仓库在两年前创建后没有更新，请问您是否成功恢复了该仓库代码？ 我再次查看查看了ClassifyAudio.ipynb中的代码，其中UrbanSound8K SVM分类是由Sklearn repo，具体是sklearn.svm.SVC[2]实现的。sklearn.svm.SVC 是基于libsvm[3]实现的，多分类支持按照一对一[1][2]方案处理的。您可以在 sklearn.svm.SVC 相关链接中找到更多信息。如果此代码仓库有帮助到您，欢迎Star，谢谢：https://www.cnblogs.com/ruidongwu/p/11869948.html [2] Sklearn.svm.SVC： https://scikit-learn.org/stable/modules/ generated/sklearn.svm.SVC。 html [3] Libsvm： https://www.csie.ntu.edu.tw/~cjlin/libsvm/ 最温暖，Leo Pan。 …… On Sun, Apr 14, 2024 at 5:59 PM moombeam1 @.> wrote: When replicating your code, I didn't understand the multi-classification method you used when using SVM in UrbanSound8K classification. Do you use one-versus-one or one-versus-rest strategy, or something else. 在复现您的代码时，我对您在UrbanSound8K 分类中使用SVM进行分类时所用的多分类方法没有看明白，请问您使用的是一对一策略还是一对多策略，亦或是其他的方法。 — Reply to this email directly, view it on GitHub <#1>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO5RI2BKI4OQD3L4UFRW4MLY5JHQVAVCNFSM6AAAAABGGDAIJCVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI2DEMBXGY2DIMY . You are receiving this because you are subscribed to this thread.Message ID: @.>

Thank you！ 感谢博主，复现成功了。