[GitHub] incubator-madlib pull request #162: MLP: Multilayer Perceptron Phase 2

2017-08-07 Thread haying
Github user haying commented on a diff in the pull request: https://github.com/apache/incubator-madlib/pull/162#discussion_r131783608 --- Diff: doc/design/modules/neural-network.tex --- @@ -22,15 +22,16 @@ \chapter{Neural Network} \begin{moduleinfo} -\item

[GitHub] incubator-madlib pull request #149: MLP: Multilayer Perceptron

2017-07-12 Thread haying
Github user haying commented on a diff in the pull request: https://github.com/apache/incubator-madlib/pull/149#discussion_r127072514 --- Diff: src/modules/convex/mlp_igd.cpp --- @@ -0,0 +1,260 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or

[GitHub] incubator-madlib pull request #149: MLP: Multilayer Perceptron

2017-07-11 Thread haying
Github user haying commented on a diff in the pull request: https://github.com/apache/incubator-madlib/pull/149#discussion_r126795536 --- Diff: src/ports/postgres/dbconnector/TypeTraits_impl.hpp --- @@ -750,6 +784,30 @@ struct TypeTraits { ); }; +template

[GitHub] incubator-madlib issue #75: SVM: Implement c++ functions for training multi-...

2016-11-14 Thread haying
Github user haying commented on the issue: https://github.com/apache/incubator-madlib/pull/75 I suggest to document these in user doc. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this

[GitHub] incubator-madlib issue #75: SVM: Implement c++ functions for training multi-...

2016-11-14 Thread haying
Github user haying commented on the issue: https://github.com/apache/incubator-madlib/pull/75 Hi Xiaocheng, Thanks for the explanation. It is good to know that it usually converges without much tuning. I would definitely try it out. --- If your project is set up for it, you can

[GitHub] incubator-madlib issue #75: SVM: Implement c++ functions for training multi-...

2016-11-14 Thread haying
Github user haying commented on the issue: https://github.com/apache/incubator-madlib/pull/75 I am not so sure about the technical details about convergence behavior of this algorithm. But it seems to me that it could be beneficial to expose the gradient so that users can do

[GitHub] incubator-madlib pull request #75: SVM: Implement c++ functions for training...

2016-11-14 Thread haying
Github user haying commented on a diff in the pull request: https://github.com/apache/incubator-madlib/pull/75#discussion_r87922091 --- Diff: src/modules/convex/algo/igd.hpp --- @@ -86,6 +101,32 @@ IGD::merge(state_type &state, template void +IGD::mergeInP

[GitHub] incubator-madlib pull request #75: SVM: Implement c++ functions for training...

2016-11-14 Thread haying
Github user haying commented on a diff in the pull request: https://github.com/apache/incubator-madlib/pull/75#discussion_r87916186 --- Diff: src/modules/convex/algo/igd.hpp --- @@ -86,6 +101,32 @@ IGD::merge(state_type &state, template void +IGD::mergeInP

[GitHub] incubator-madlib pull request: SVM: Add Gaussian kernel feature ma...

2015-12-24 Thread haying
Github user haying commented on a diff in the pull request: https://github.com/apache/incubator-madlib/pull/10#discussion_r48407523 --- Diff: methods/array_ops/src/pg_gp/array_ops.sql_in --- @@ -481,6 +481,18 @@ m4_ifdef(`__HAS_FUNCTION_PROPERTIES__', `NO SQL', `');