05:58 < travis-ci> Prabhat-IIT/mlpack#10 (SAGA - 51f3943 : Prabhat): The build has errored.

05:58 < travis-ci> Change view : https://github.com/Prabhat-IIT/mlpack/compare/f29849154ece...51f3943fa059

05:58 < travis-ci> Build details : https://travis-ci.org/Prabhat-IIT/mlpack/builds/364443523

06:03 < manthan> rcurtin : in relation to the decision tree pruning implementation, i have figured out that if classProbabilities for leaf nodes is made to hold the count of points for each class insted of probabilities, then there will be no need to include any additional data members

06:05 < manthan> additional data members are just needed to figure the majority class of a particular non-leaf node. But if counts are stored, then this can be found by just looping over the classProbabilities of each child of the current non-leaf node.

12:15 < travis-ci> Prabhat-IIT/mlpack#11 (SAGA - 494035e : Prabhat): The build has errored.

12:15 < travis-ci> Change view : https://github.com/Prabhat-IIT/mlpack/compare/51f3943fa059...494035efda1f

12:15 < travis-ci> Build details : https://travis-ci.org/Prabhat-IIT/mlpack/builds/364558964

14:22 < rajiv_> This is a continuation of the discussion on 8th April in the IRC. I was able to run the tests using the command: bin/mlpack_test. But, I still get the error: use of LAPACK must be enabled. To check if I have LAPACK installed, I tried building armadillo. When I ran cmake, it said that I have LAPACK installed. I am not able to figure out the reason for this issue.

14:25 < zoq> rajiv_: Can't you use the package manager to install armadillo?

14:26 < rajiv_> I have armadillo installed already... I just tried to build it to check if I have LAPACK or not

14:37 < rajiv_> zoq_: This is the error I am getting: https://imagebin.ca/v/3xucl1iet744

16:24 < pryan> hello

16:24 < pryan> how can i implement mlpack into a C++ code to get rid of spam of mi ircd ( im codding an ircd )

16:24 < pryan> what functions should i use ?

16:41 < luffy1996> @zoq I am presently going through asynchronous methods for deep RL implementation on mlpack. If I remember correctly then you mentioned that mlpack supports implementation of A3C (Asynchronous advantage actor critic). However I do not find the sources of the file. I can see only three of the 4 variants discussed in the paper implemented . I might be wrong here but would you please give me some lead.

16:41 < pryan> hello luffy1996

16:41 < pryan> how can i implement mlpack into a C++ code to get rid of spam of mi ircd ( im codding an ircd )

16:44 < luffy1996> @zoq Also I am stuck with policy gradient application in mlpack . How should I proceed with gradient ascent in mlpack.

16:46 < luffy1996> @pryan I do not think I could help you in this. Thanks :)

16:51 < pryan> ok

16:52 < travis-ci> Prabhat-IIT/mlpack#12 (fastmksTest - 2df6716 : Prabhat): The build has errored.

16:52 < travis-ci> Change view : https://github.com/Prabhat-IIT/mlpack/compare/8a3fd13cea1c...2df6716bdcc3

16:52 < travis-ci> Build details : https://travis-ci.org/Prabhat-IIT/mlpack/builds/364684477

17:11 < luffy1996> \

17:12 < travis-ci> Prabhat-IIT/mlpack#13 (fastmksTest - 795dca3 : Prabhat): The build has errored.

17:12 < travis-ci> Change view : https://github.com/Prabhat-IIT/mlpack/compare/2df6716bdcc3...795dca3ebe44

17:12 < travis-ci> Build details : https://travis-ci.org/Prabhat-IIT/mlpack/builds/364694449

17:42 < travis-ci> Prabhat-IIT/mlpack#14 (fastmksTest - d547e4d : Prabhat): The build has errored.

17:42 < travis-ci> Change view : https://github.com/Prabhat-IIT/mlpack/compare/795dca3ebe44...d547e4d6b8dc

17:42 < travis-ci> Build details : https://travis-ci.org/Prabhat-IIT/mlpack/builds/364696191

21:09 < zoq> rajiv_: Okay, can you tell me more about your OS (ubuntu)?

21:09 < pryan> zoq hello

21:10 < pryan> there is an example of how can i implement ML ?

21:10 < pryan> im makin my own ircd

21:10 < pryan> and want to have ML to do spam stuff

21:11 < zoq> luffy1996: hm, A3C is implemented, there is an open issue: https://github.com/mlpack/mlpack/pull/1091 which talks about an Alias layer, which is required to implement A3C.

21:17 < zoq> pryan: I think, there are multiple ideas you could test, nearest neighbor classifier, logistic regression, neural networks.

21:17 < pryan> hm

21:18 < pryan> some example of code or anything ?

21:18 < pryan> a wiki

21:18 < pryan> or some

21:19 < zoq> Unfortunately mlpack doesn't provide a word to vector method; which I think is the first step in the pipeline.

21:19 < pryan> aham

21:20 < zoq> See: http://www.mlpack.org/docs/mlpack-git/doxygen/tutorials.html and I guess another good resource are the tests, which in some cases provide simple examples that could be directly used: https://github.com/mlpack/mlpack/tree/master/src/mlpack/tests

21:20 < pryan> lot of thanks zoq

21:20 < pryan> your job are great

21:21 < zoq> pryan: Here to help, let me know if I should clarify anything.

21:21 < pryan> zoq i have read a little about ML

21:21 < pryan> i think that load from csv is out

21:22 < pryan> so probably i must write some scruppy stuff

21:22 < pryan> and again, thanks

21:24 < zoq> yeah, depending on the input some boilerplate code is necessary, but I think some standard C++ functions is all you need ... not sure.

21:24 < pryan> ok

22:08 < travis-ci> Prabhat-IIT/mlpack#15 (newinit - cb1d2d4 : Prabhat): The build has errored.

22:08 < travis-ci> Change view : https://github.com/Prabhat-IIT/mlpack/compare/96480b97a85e...cb1d2d4b88fa

22:08 < travis-ci> Build details : https://travis-ci.org/Prabhat-IIT/mlpack/builds/364824106