tag:blogger.com,1999:blog-7345806147365425073.post4380410012117103674..comments2019-07-19T10:53:02.106+02:00Comments on Peekaboo: Update for structured SVM in PythonAndreas Muellerhttp://www.blogger.com/profile/10177962095394942563noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-7345806147365425073.post-4830468637574314932012-06-18T00:02:50.163+02:002012-06-18T00:02:50.163+02:00To elaborate: I am not very interested in the opti...To elaborate: I am not very interested in the optimization problem itself. I want to learn in a setting with approximate inference, where the inference is potentially much, much more expensive than the SVM optimization. As I said in my last post, I even start the SVM optimization completely anew in each iteration. That doesn't really matter that much.<br />The idea was more to have a flexible, easy to understand system. This is in no way an approach to be faster than the established software.Andreas Muellerhttps://www.blogger.com/profile/10177962095394942563noreply@blogger.comtag:blogger.com,1999:blog-7345806147365425073.post-8219983950387792932012-06-17T23:55:51.858+02:002012-06-17T23:55:51.858+02:00No, I use one slack per example. This is the stand...No, I use one slack per example. This is the standard version from the original paper.Andreas Muellerhttps://www.blogger.com/profile/10177962095394942563noreply@blogger.comtag:blogger.com,1999:blog-7345806147365425073.post-23860410971659254482012-06-17T23:54:30.870+02:002012-06-17T23:54:30.870+02:00> Therefore, the sum of all alpha that correspo...> Therefore, the sum of all alpha that correspond to a single example is bound by C.<br /><br />Donâ€™t you implement the 1-slack variant?Roman Shapovalovhttps://www.blogger.com/profile/01422787855469629259noreply@blogger.com