Re: [mlpack] GSOC 2021 project ideas

[Anmolpreet Singh]

Hello Ryan and Marcus, Thanks for the feedback.   I agree with Ryan that it is necessary to document with examples the methods which are added and I ensure that it will be in the plan and also to the point that a list of metrics can be implemented. I have mentioned some metrics(SSIM, PSNR, Top K Accuracy) in the previous mail which I have gone through. While searching for metrics I  came across AUC (Area under ROC curve) as a classification evaluation metric. Everyone is welcome to add a metric to the above ones.   Marcus, I have also gone through VMAF and it is an open source software developed by Netflix which uses several image quality metrics into a ML model to calculate video quality to evaluate encoding method with maximum quality retained. This is also motivated from image quality metrics.   I would always love to hear suggestions on Useful metrics from the community.   Regards, Anmolpreet Singh   From: Marcus Edel Sent: 15 March 2021 21:29 To: Ryan Curtin Cc: Anmolpreet Singh; mlpack@lists.mlpack.org Subject: Re: [mlpack] GSOC 2021 project ideas   Hello Anmolpreet,   I like the idea, also are you aware of https://github.com/Netflix/vmaf? I really like what Netflix put together.   Thanks, Marcus On 13. Mar 2021, at 17:16, Ryan Curtin <r...@ratml.org> wrote:   Hi Anmolpreet, These sound like nice ideas.  It seems likely to me that it would be possible to implement more than one of these in the summer, but also you should be sure to think about example usages: if we have many different quality metrics, that is great, but what is maybe more important to drive users to them is coherent and clear examples demonstrating their usage.  So you may want to ensure that documentation and examples are part of the plan also. :) I hope this is helpful!  I'm not deeply involved with #2294, so others may have other comments too. Thanks! Ryan On Sat, Mar 13, 2021 at 11:55:22PM +0530, Anmolpreet Singh wrote:   Hello all!       Being with MLpack community for some time I realized a few ideas which can   be implemented during period of GSOC 2021. Taking into consideration the   shorter time this year I want to propose the following idea of   implementing  some useful metrics of ML as a GSOC 2021 summer project.       The structural similarity index measure (SSIM) is a method for predicting   the perceived quality of images or measuring the similarity between 2   images considering the structural information. It has good applications in   image compression (checking quality of compressed image), image   restoration and pattern recognition. I think it will be some interesting   stuff to add in MLpack, as it is an important part of image processing.   Also, its need is highlighted from the issue #2294(Addition of essential   metrics only) which is pending due to issues in implementation. So, it   will be good if is done in an organized way under guidance of mentors. I   have also gone through a couple of recent research papers regarding   improvements in this.       PSNR is another image quality metric which has wide use in digital image   processing. I feel that Metrics like these which have tremendous use will   fit good in MLpack. Also, discussion will bring more metrics into   consideration.       In addition, I may also add metrics like Top K accuracy and some other   needful metrics (suggestions are welcomed) which may help in evaluating   the performance of various models. Recently,  I have done a data science   project (chat-bot using NLP)  and planning to continue my journey with   this idea.       Any feedback or suggestion for these ideas will be really helpful for   further planning this according to the discussion.   Regards,   Anmolpreet Singh     _______________________________________________ mlpack mailing list mlpack@lists.mlpack.org http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack -- Ryan Curtin    | "We need some time for some things to happen!" r...@ratml.org |   - Bells _______________________________________________ mlpack mailing list mlpack@lists.mlpack.org http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack

_______________________________________________
mlpack mailing list
mlpack@lists.mlpack.org
http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack