[jira] [Commented] (SINGA-253) Net converter for caffe model
[ https://issues.apache.org/jira/browse/SINGA-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15670428#comment-15670428 ] pangxiell commented on SINGA-253: - Have you supported proto.txt with trained.caffemodel which contain trained value and having much meanings for transform platform and transform learning(something very useful like MXNET https://github.com/dmlc/mxnet/tree/master/tools/caffe_converter) ? In a convert work, most important thing is the trained value. > Net converter for caffe model > - > > Key: SINGA-253 > URL: https://issues.apache.org/jira/browse/SINGA-253 > Project: Singa > Issue Type: New Feature >Reporter: Xiangrui >Assignee: Xiangrui > > Convert caffe model into singa model. > It is a very basic implementation. Now it can convert feed forward net and > supports pysinga. > Implementation method: > 1. read proto file of caffe model using caffe.proto, and serialize it to > string. > 2. parse the string use singa model.proto and get the layer config. > 3. setup each layer and add it to a feed forward net. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (SINGA-253) Net converter for caffe model
[ https://issues.apache.org/jira/browse/SINGA-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15593903#comment-15593903 ] Xiangrui commented on SINGA-253: Thanks for your comments! We are working on converting trained models now. Please check our github repository later. > Net converter for caffe model > - > > Key: SINGA-253 > URL: https://issues.apache.org/jira/browse/SINGA-253 > Project: Singa > Issue Type: New Feature >Reporter: Xiangrui > > Convert caffe model into singa model. > It is a very basic implementation. Now it can convert feed forward net and > supports pysinga. > Implementation method: > 1. read proto file of caffe model using caffe.proto, and serialize it to > string. > 2. parse the string use singa model.proto and get the layer config. > 3. setup each layer and add it to a feed forward net. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (SINGA-253) Net converter for caffe model
[ https://issues.apache.org/jira/browse/SINGA-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15593688#comment-15593688 ] pangxiell commented on SINGA-253: - Will you support proto.txt with trained.caffemodel which contain trained value and having much meanings for transform platform and transform learning ? In a convert work, most important thing is the traned value. > Net converter for caffe model > - > > Key: SINGA-253 > URL: https://issues.apache.org/jira/browse/SINGA-253 > Project: Singa > Issue Type: New Feature >Reporter: Xiangrui > > Convert caffe model into singa model. > It is a very basic implementation. Now it can convert feed forward net and > supports pysinga. > Implementation method: > 1. read proto file of caffe model using caffe.proto, and serialize it to > string. > 2. parse the string use singa model.proto and get the layer config. > 3. setup each layer and add it to a feed forward net. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (SINGA-253) Net converter for caffe model
[ https://issues.apache.org/jira/browse/SINGA-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15585413#comment-15585413 ] ASF subversion and git services commented on SINGA-253: --- Commit 77405ec19565ed82feedb933cf562f61c44522c4 in incubator-singa's branch refs/heads/master from [~Xiangrui] [ https://git-wip-us.apache.org/repos/asf?p=incubator-singa.git;h=77405ec ] SINGA-253 Net converter for caffe model Convert caffe model into singa model. It is a very basic implementation. Now it can convert feed forward net and supports pysinga. Implementation method: 1. read proto file of caffe model using caffe.proto, and serialize it to string. 2. parse the string use singa model.proto and get the layer config. 3. setup each layer and add it to a feed forward net. Update and example: 1. put conversion funcions into layer.py and converter.py; 2. add 'caffe' option in cifar example which converts alexnet from caffe to singa. > Net converter for caffe model > - > > Key: SINGA-253 > URL: https://issues.apache.org/jira/browse/SINGA-253 > Project: Singa > Issue Type: New Feature >Reporter: Xiangrui > > Convert caffe model into singa model. > It is a very basic implementation. Now it can convert feed forward net and > supports pysinga. > Implementation method: > 1. read proto file of caffe model using caffe.proto, and serialize it to > string. > 2. parse the string use singa model.proto and get the layer config. > 3. setup each layer and add it to a feed forward net. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (SINGA-253) Net converter for caffe model
[ https://issues.apache.org/jira/browse/SINGA-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15585414#comment-15585414 ] ASF subversion and git services commented on SINGA-253: --- Commit 32b4368f07b4c55880fa04bcc2d9f39ab9d75fc5 in incubator-singa's branch refs/heads/master from [~Xiangrui] [ https://git-wip-us.apache.org/repos/asf?p=incubator-singa.git;h=32b4368 ] SINGA-253 Net converter for caffe model Modification of pooling layer add a 'ceil' field in proto to be compatible with caffe train alexnet from caffe, achieving 82% accuracy. > Net converter for caffe model > - > > Key: SINGA-253 > URL: https://issues.apache.org/jira/browse/SINGA-253 > Project: Singa > Issue Type: New Feature >Reporter: Xiangrui > > Convert caffe model into singa model. > It is a very basic implementation. Now it can convert feed forward net and > supports pysinga. > Implementation method: > 1. read proto file of caffe model using caffe.proto, and serialize it to > string. > 2. parse the string use singa model.proto and get the layer config. > 3. setup each layer and add it to a feed forward net. -- This message was sent by Atlassian JIRA (v6.3.4#6332)