Show HN: Aurora- a comprehensive set of Statistics and Machine Learning tools | Hacker News Reader

Show HN: Aurora- a comprehensive set of Statistics and Machine Learning tools

github.com

11 pointsmariusnea9 comments

I can’t actually figure out why anyone would want to use this. The readme is pretty vague and doesn’t give me any setup instructions. More importantly, I don’t understand the advantage of this over the existing tools.

Hello, and thank you for your time. Considering its current features and focus, Aurora seems more tailored towards beginners. Practically, in less than 5 minutes you can have a working model applied to your data without the need for a steep learning curve. Another advantage is that Aurora is not limited in terms of the number of lines and columns in which the data is divided. Looking into the future, the plugin based system will give it an advantage of a rapid development of new functionalities without the need to make changes in the base code. Also the future possibility to connect with various data sources will be an advantage.

Is this generally an advertisement to drive traffic to the personal blog?

It doesn't seem to be a comprehensive set of stat and ML tools.

No, of course not. Why would I put Github insted of my blog, as a link? The only reason that I've put my blog URL in the README file from Github is that people wanted a download and run version. And for some reason I can not upload to Github a file that is more than 100MB in size. Regarding the options present in AURORA, notice that is still in its infancy. Some of them are present, others not. But probably more important is that Aurora offers the posibility that these options can be easily added through plugins. We will see in an year how it will look. Thank you for your time.

I'm sure you've got some cool stuff here, but until you update your readme with at minimum one full, self contained working example, no-one is going to try this.

You should also turn your readme into a markdown file and incorporate the HOWTO into it. It'll make it appear a lot friendlier.

Yes, now I realize that it was more adequate to make the interaction with the product easier at first glance. Perhaps a presentation video would have been welcome. Thank you for the advice.

No examples (code), no screenshots, no comparison with existing tools, no feature list ... no thanks.

What even is "Optimized Research Analytics"?

Hello, thank you for your reply. There is an working executable which can be downloaded from my personal website. I omitted to make it more visible. Now I've added a link in the Readme file from repository.

Regarding your last question, in the world of data science there are three types of analytics, descriptive, predictive and prescriptive. Prescriptive analytics techniques, like optimization, tell you what actions to take to achieve a desired outcome. So after descriptive and predictive analytics research activities are done, their outputs become inputs for optimization. After optimization research activity is done we have a complete solution for our problem.

Reply on news.ycombinator.com