[jira] [Commented] (MADLIB-1134) Neural Networks - MLP - Phase 2
[ https://issues.apache.org/jira/browse/MADLIB-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105853#comment-16105853 ] Frank McQuillan commented on MADLIB-1134: - That looks like a good model to check against. > Neural Networks - MLP - Phase 2 > --- > > Key: MADLIB-1134 > URL: https://issues.apache.org/jira/browse/MADLIB-1134 > Project: Apache MADlib > Issue Type: Improvement > Components: Module: Neural Networks >Reporter: Frank McQuillan >Assignee: Cooper Sloan > Fix For: v1.12 > > > Follow on from https://issues.apache.org/jira/browse/MADLIB-413 > Story > As a MADlib developer, I want to get 2nd phase implementation of NN going > with training and prediction functions, so that I can use this to build to an > MVP version for GA. > Features to add: > * weights for inputs > * logic for n_tries > * normalize inputs > * L2 regularization > * learning rate policy > * warm start -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Comment Edited] (MADLIB-1134) Neural Networks - MLP - Phase 2
[ https://issues.apache.org/jira/browse/MADLIB-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105769#comment-16105769 ] Cooper Sloan edited comment on MADLIB-1134 at 7/28/17 10:06 PM: Let's do something like this: [https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neural_network/tests/test_mlp.py#L79 for ensuring correctness of our backprop. was (Author: coopersloan): Let's do something like [https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neural_network/tests/test_mlp.py#L79](this) for ensuring correctness of our backprop. > Neural Networks - MLP - Phase 2 > --- > > Key: MADLIB-1134 > URL: https://issues.apache.org/jira/browse/MADLIB-1134 > Project: Apache MADlib > Issue Type: Improvement > Components: Module: Neural Networks >Reporter: Frank McQuillan >Assignee: Cooper Sloan > Fix For: v1.12 > > > Follow on from https://issues.apache.org/jira/browse/MADLIB-413 > Story > As a MADlib developer, I want to get 2nd phase implementation of NN going > with training and prediction functions, so that I can use this to build to an > MVP version for GA. > Features to add: > * weights for inputs > * logic for n_tries > * normalize inputs > * L2 regularization > * learning rate policy > * warm start -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Comment Edited] (MADLIB-1134) Neural Networks - MLP - Phase 2
[ https://issues.apache.org/jira/browse/MADLIB-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105769#comment-16105769 ] Cooper Sloan edited comment on MADLIB-1134 at 7/28/17 10:06 PM: Let's do something like this: https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neural_network/tests/test_mlp.py#L79 for ensuring correctness of our backprop. was (Author: coopersloan): Let's do something like this: [https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neural_network/tests/test_mlp.py#L79 for ensuring correctness of our backprop. > Neural Networks - MLP - Phase 2 > --- > > Key: MADLIB-1134 > URL: https://issues.apache.org/jira/browse/MADLIB-1134 > Project: Apache MADlib > Issue Type: Improvement > Components: Module: Neural Networks >Reporter: Frank McQuillan >Assignee: Cooper Sloan > Fix For: v1.12 > > > Follow on from https://issues.apache.org/jira/browse/MADLIB-413 > Story > As a MADlib developer, I want to get 2nd phase implementation of NN going > with training and prediction functions, so that I can use this to build to an > MVP version for GA. > Features to add: > * weights for inputs > * logic for n_tries > * normalize inputs > * L2 regularization > * learning rate policy > * warm start -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (MADLIB-1134) Neural Networks - MLP - Phase 2
[ https://issues.apache.org/jira/browse/MADLIB-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105769#comment-16105769 ] Cooper Sloan commented on MADLIB-1134: -- Let's do something like [this](https://issues.apache.org/jira/browse/MADLIB-1134) for ensuring correctness of our backprop. > Neural Networks - MLP - Phase 2 > --- > > Key: MADLIB-1134 > URL: https://issues.apache.org/jira/browse/MADLIB-1134 > Project: Apache MADlib > Issue Type: Improvement > Components: Module: Neural Networks >Reporter: Frank McQuillan >Assignee: Cooper Sloan > Fix For: v1.12 > > > Follow on from https://issues.apache.org/jira/browse/MADLIB-413 > Story > As a MADlib developer, I want to get 2nd phase implementation of NN going > with training and prediction functions, so that I can use this to build to an > MVP version for GA. > Features to add: > * weights for inputs > * logic for n_tries > * normalize inputs > * L2 regularization > * learning rate policy > * warm start -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Comment Edited] (MADLIB-1134) Neural Networks - MLP - Phase 2
[ https://issues.apache.org/jira/browse/MADLIB-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105769#comment-16105769 ] Cooper Sloan edited comment on MADLIB-1134 at 7/28/17 10:05 PM: Let's do something like [https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neural_network/tests/test_mlp.py#L79](this) for ensuring correctness of our backprop. was (Author: coopersloan): Let's do something like [this](https://issues.apache.org/jira/browse/MADLIB-1134) for ensuring correctness of our backprop. > Neural Networks - MLP - Phase 2 > --- > > Key: MADLIB-1134 > URL: https://issues.apache.org/jira/browse/MADLIB-1134 > Project: Apache MADlib > Issue Type: Improvement > Components: Module: Neural Networks >Reporter: Frank McQuillan >Assignee: Cooper Sloan > Fix For: v1.12 > > > Follow on from https://issues.apache.org/jira/browse/MADLIB-413 > Story > As a MADlib developer, I want to get 2nd phase implementation of NN going > with training and prediction functions, so that I can use this to build to an > MVP version for GA. > Features to add: > * weights for inputs > * logic for n_tries > * normalize inputs > * L2 regularization > * learning rate policy > * warm start -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Closed] (MADLIB-1101) Graph - weakly connected components helper functions
[ https://issues.apache.org/jira/browse/MADLIB-1101?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Frank McQuillan closed MADLIB-1101. --- > Graph - weakly connected components helper functions > > > Key: MADLIB-1101 > URL: https://issues.apache.org/jira/browse/MADLIB-1101 > Project: Apache MADlib > Issue Type: New Feature > Components: Module: Graph >Reporter: Frank McQuillan > Fix For: v1.12 > > > Context > Follow on from > https://issues.apache.org/jira/browse/MADLIB-1071 > Story > As a data scientist, I want to use helper functions related to weakly > connected components, so that I don't have to query the result table myself > which is less efficient and subject to error. > List of helper functions roughly in priority order: > 1) biggest connected component > 2) number of nodes per connected component (histogram) > 3) whether two nodes belong to same or different connected components > 4) count of connected cpt clusters > 5) Set of all nodes which can be reached (have a path) from a specified vertex -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (MADLIB-1101) Graph - weakly connected components helper functions
[ https://issues.apache.org/jira/browse/MADLIB-1101?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105627#comment-16105627 ] ASF GitHub Bot commented on MADLIB-1101: Github user asfgit closed the pull request at: https://github.com/apache/incubator-madlib/pull/155 > Graph - weakly connected components helper functions > > > Key: MADLIB-1101 > URL: https://issues.apache.org/jira/browse/MADLIB-1101 > Project: Apache MADlib > Issue Type: New Feature > Components: Module: Graph >Reporter: Frank McQuillan > Fix For: v1.12 > > > Context > Follow on from > https://issues.apache.org/jira/browse/MADLIB-1071 > Story > As a data scientist, I want to use helper functions related to weakly > connected components, so that I don't have to query the result table myself > which is less efficient and subject to error. > List of helper functions roughly in priority order: > 1) biggest connected component > 2) number of nodes per connected component (histogram) > 3) whether two nodes belong to same or different connected components > 4) count of connected cpt clusters > 5) Set of all nodes which can be reached (have a path) from a specified vertex -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (MADLIB-1101) Graph - weakly connected components helper functions
[ https://issues.apache.org/jira/browse/MADLIB-1101?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16105503#comment-16105503 ] Frank McQuillan commented on MADLIB-1101: - 1) Please check validity of vertex inputs in graph_wcc_vertex_check graph_wcc_reachable_vertices and give a nice error message if an invalid vertex is entered. Currently this is not trapped. 2) I posted an updated WCC notebook to the github folder https://github.com/apache/incubator-madlib-site/blob/asf-site/community-artifacts/Weakly-connected-cpts-v2.ipynb that includes the helper functions from this JIRA > Graph - weakly connected components helper functions > > > Key: MADLIB-1101 > URL: https://issues.apache.org/jira/browse/MADLIB-1101 > Project: Apache MADlib > Issue Type: New Feature > Components: Module: Graph >Reporter: Frank McQuillan > Fix For: v1.12 > > > Context > Follow on from > https://issues.apache.org/jira/browse/MADLIB-1071 > Story > As a data scientist, I want to use helper functions related to weakly > connected components, so that I don't have to query the result table myself > which is less efficient and subject to error. > List of helper functions roughly in priority order: > 1) biggest connected component > 2) number of nodes per connected component (histogram) > 3) whether two nodes belong to same or different connected components > 4) count of connected cpt clusters > 5) Set of all nodes which can be reached (have a path) from a specified vertex -- This message was sent by Atlassian JIRA (v6.4.14#64029)
