Hello Trevor, I have gone through the two links that you sent me. Although I am not familiar with scala, I was able to figure out that the files Fitter.scala, Model.scala, UnsupervisedFitter.scala contain traits (which are similar to interfaces in Java) and the LinearRegressionModel.scala contains the core code for regression. I wasn't able to understand specific syntactic terms like *trait LinearRegressionModel[K] extends RegressionModel[K]*, what is K here?
With respect to my knowledge in Scala, I've never had the opportunity to learn / work in Scala but I got a sense that it's model is similar to that of Java. Having worked in Java, I could see some basic similarities in both the languages' models. I've read that Scala is a language where OOP meets the functional paradigm. Also, Could you let me know where I could find the list of algorithms that Mahout implements along with its documentation and what algorithms are planned to be implemented soon? The main web page just lists down the names. Thanks, Aditya On Thu, Feb 23, 2017 at 6:57 PM, Trevor Grant <trevor.d.gr...@gmail.com> wrote: > Hey Aditya- > > First of all, welcome to the community. We'd love to have you help > contribute. > > The new algorithms framework is certainly a 'target rich environment'. > > Since you already are familiar with DBSCAN, why not start there? > > If you check out: > https://github.com/apache/mahout/tree/master/math-scala/ > src/main/scala/org/apache/mahout/math/algorithms > > You'll see in general what our framework looks like.. > > You'll need to create a ClassificationModel trait similar to: > https://github.com/apache/mahout/blob/master/math-scala/ > src/main/scala/org/apache/mahout/math/algorithms/ > regression/RegressorModel.scala > > Then you'll extend the the ClassificationModel with DBSCAN (or possibly > some intermediate trait, as LinearRegressionModel does before OLS). > > Perhaps I should have started by asking- how well do you know scala? > > Anyway, those are good places to get started! Let me know if I can help. > > tg > > Trevor Grant > Data Scientist > https://github.com/rawkintrevo > http://stackexchange.com/users/3002022/rawkintrevo > http://trevorgrant.org > > *"Fortunate is he, who is able to know the causes of things." -Virgil* > > > On Wed, Feb 22, 2017 at 4:23 PM, Aditya <adityasarma...@gmail.com> wrote: > > > Hello everyone! > > > > I'm a senior year computer science student from Birla Institute of > > Technology and Science, India. I have experience in fields like Data > Mining > > and Machine Learning. Apart from doing basic coursework which included > Data > > Mining, Parallel Computing, and Machine Learning I have also worked on > > research projects where I worked on building scalable DBSCAN like > > clustering algorithms. > > > > I have gone through the Apache Mahout website and was wondering if I > > could *contribute > > to Mahout in terms of algorithm **development / improvising existing > > algorithms.* > > > > I would be grateful if you could provide me with a starting point, from > > where I can pick up and understand the Mahout ecosystem. I have no > previous > > experience in working with Apache Mahout or Spark but I have worked with > > the Map reduce model before (but haven't used Hadoop) > > > > I wish to work full time during summer and take part in the Google Summer > > of Code 2017 program by contributing to Apache Mahout. > > > > > > Awaiting your replies! > > > > Cheers! > > Aditya > > >
