Closed exalate-issue-sync[bot] closed 1 year ago
exalate-issue-sync[bot]
commented
1 year ago Avkash Chauhan commented: [~accountid:557058:389d9607-5bd8-4611-8c6a-755fe9295223] [[~accountid:557058:3bc534f4-c129-4d5f-b8c1-5a69d34942ee] What do you guys suggest regarding adding this feature?
exalate-issue-sync[bot]
commented
1 year ago Arno Candel commented: That would be GLM...
exalate-issue-sync[bot]
commented
1 year ago Navdeep commented: Regarding http://ufldl.stanford.edu/wiki/index.php/Stacked_Autoencoders under Concrete Examples: I think there might be some confusion here. The first two diagrams are autoencoders in which the first feeds into the next. The third diagram is not really a separate network. Just a zoom in of the fourth diagram. So, you should be able to get the features 1,2 and use a softmax classifier from figure 4
exalate-issue-sync[bot]
commented
1 year ago Nidhi Mehta commented: works now - https://github.com/h2oai/h2o-3/blob/master/h2o-r/tests/testdir_algos/deeplearning/runit_deeplearning_no_hidden.R runit - here is the syntax -
{code:java} hh3 <- h2o.deeplearning(training_frame=train_w_resp, x=1:(ncol(train_w_resp)-1), y=ncol(train_w_resp), hidden=numeric(), epochs=1,export_weights_and_biases = T)
{code}
h2o-ops
commented
1 year ago JIRA Issue Migration Info
Jira Issue: PUBDEV-4364 Assignee: Arno Candel Reporter: Nidhi Mehta State: Resolved Fix Version: N/A Attachments: N/A Development PRs: Available
Linked PRs from JIRA
90658 and #90664
http://ufldl.stanford.edu/wiki/index.php/Stacked_Autoencoders There is no way to set hidden layer to zero to connect input directly to output in h2o deep learning
{code:java} TRAIN <- "/Users/nidhimehta/train.csv.gz" TEST <- "/Users/nidhimehta/test.csv.gz" response <- 785 train_hex <- h2o.importFile(TRAIN) test_hex <- h2o.importFile(TEST )
train <- train_hex[,-response] test <- test_hex [,-response] train_hex[,response] <- as.factor(train_hex[,response]) test_hex [,response] <- as.factor(test_hex [,response])
hh3 <- h2o.deeplearning(training_frame=train_hex, x=1:(ncol(train_hex)-1), y=ncol(train_hex), hidden=c(0), epochs=1,export_weights_and_biases = T)
ERROR: Unexpected HTTP Status code: 412 Precondition Failed (url = http://localhost:54321/3/ModelBuilders/deeplearning)
water.exceptions.H2OModelBuilderIllegalArgumentException [1] "water.exceptions.H2OModelBuilderIllegalArgumentException: Illegal argument(s) for DeepLearning model: DeepLearning_model_R_1493341612082_2. Details: ERRR on field: _hidden: Hidden layer size must be positive.\n" [2] " water.exceptions.H2OModelBuilderIllegalArgumentException.makeFromBuilder(H2OModelBuilderIllegalArgumentException.java:20)"
[3] " hex.ModelBuilder.trainModel(ModelBuilder.java:198)"
[4] " water.api.ModelBuilderHandler.handle(ModelBuilderHandler.java:52)"
[5] " water.api.ModelBuilderHandler.handle(ModelBuilderHandler.java:16)"
[6] " water.api.RequestServer.serve(RequestServer.java:448)"
[7] " water.api.RequestServer.doGeneric(RequestServer.java:297)"
[8] " water.api.RequestServer.doPost(RequestServer.java:223)"
[9] " javax.servlet.http.HttpServlet.service(HttpServlet.java:755)"
[10] " javax.servlet.http.HttpServlet.service(HttpServlet.java:848)"
[11] " org.eclipse.jetty.servlet.ServletHolder.handle(ServletHolder.java:684)"
[12] " org.eclipse.jetty.servlet.ServletHandler.doHandle(ServletHandler.java:503)"
[13] " org.eclipse.jetty.server.handler.ContextHandler.doHandle(ContextHandler.java:1086)"
[14] " org.eclipse.jetty.servlet.ServletHandler.doScope(ServletHandler.java:429)"
[15] " org.eclipse.jetty.server.handler.ContextHandler.doScope(ContextHandler.java:1020)"
[16] " org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:135)"
[17] " org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:154)"
[18] " org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:116)"
[19] " water.JettyHTTPD$LoginHandler.handle(JettyHTTPD.java:418)"
[20] " org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:154)"
[21] " org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:116)"
[22] " org.eclipse.jetty.server.Server.handle(Server.java:370)"
[23] " org.eclipse.jetty.server.AbstractHttpConnection.handleRequest(AbstractHttpConnection.java:494)"
[24] " org.eclipse.jetty.server.BlockingHttpConnection.handleRequest(BlockingHttpConnection.java:53)"
[25] " org.eclipse.jetty.server.AbstractHttpConnection.content(AbstractHttpConnection.java:982)"
[26] " org.eclipse.jetty.server.AbstractHttpConnection$RequestHandler.content(AbstractHttpConnection.java:1043)"
[27] " org.eclipse.jetty.http.HttpParser.parseNext(HttpParser.java:865)"
[28] " org.eclipse.jetty.http.HttpParser.parseAvailable(HttpParser.java:240)"
[29] " org.eclipse.jetty.server.BlockingHttpConnection.handle(BlockingHttpConnection.java:72)"
[30] " org.eclipse.jetty.server.bio.SocketConnector$ConnectorEndPoint.run(SocketConnector.java:264)"
[31] " org.eclipse.jetty.util.thread.QueuedThreadPool.runJob(QueuedThreadPool.java:608)"
[32] " org.eclipse.jetty.util.thread.QueuedThreadPool$3.run(QueuedThreadPool.java:543)"
[33] " java.lang.Thread.run(Thread.java:744)"
Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = page, :
ERROR MESSAGE:
Illegal argument(s) for DeepLearning model: DeepLearning_model_R_1493341612082_2. Details: ERRR on field: _hidden: Hidden layer size must be positive.
{code}