Error in training cat-dog classification model as in example.

[Tapojit Debnath]

Hi, I have been attempting to train the cat dog classification model as in the cortex example, but I am facing an error saying "Address already in use". Here is the error stacktrace:

Failed trying to require cortex.training with: java.net.BindException: Address already in use
                 (Unknown Source) sun.nio.ch.Net.bind0
                     Net.java:433 sun.nio.ch.Net.bind
                     Net.java:425 sun.nio.ch.Net.bind
 ServerSocketChannelImpl.java:223 sun.nio.ch.ServerSocketChannelImpl.bind
      ServerSocketAdaptor.java:74 sun.nio.ch.ServerSocketAdaptor.bind
      ServerSocketAdaptor.java:67 sun.nio.ch.ServerSocketAdaptor.bind
               HttpServer.java:77 org.httpkit.server.HttpServer.<init>
                    server.clj:42 org.httpkit.server/run-server
                     server.clj:8 org.httpkit.server/run-server
                  RestFn.java:423 clojure.lang.RestFn.invoke
                     core.clj:138 think.gate.core/open
                     core.clj:120 think.gate.core/open
                  RestFn.java:559 clojure.lang.RestFn.invoke
           classification.clj:222 cortex.experiment.classification/display-dataset-and-model
           classification.clj:211 cortex.experiment.classification/display-dataset-and-model
           classification.clj:245 cortex.experiment.classification/create-listener[fn]
           classification.clj:308 cortex.experiment.classification/perform-experiment
           classification.clj:302 cortex.experiment.classification/perform-experiment
                  training.clj:63 cortex.training/eval64115
                  training.clj:60 cortex.training/eval64115
               Compiler.java:6927 clojure.lang.Compiler.eval
               Compiler.java:7379 clojure.lang.Compiler.load
                      RT.java:372 clojure.lang.RT.loadResourceScript
                      RT.java:363 clojure.lang.RT.loadResourceScript
                      RT.java:453 clojure.lang.RT.load
                      RT.java:419 clojure.lang.RT.load
                    core.clj:5893 clojure.core/load[fn]
                    core.clj:5892 clojure.core/load
                    core.clj:5876 clojure.core/load
                  RestFn.java:408 clojure.lang.RestFn.invoke
                    core.clj:5697 clojure.core/load-one
                    core.clj:5692 clojure.core/load-one
                    core.clj:5737 clojure.core/load-lib[fn]
                    core.clj:5736 clojure.core/load-lib
                    core.clj:5717 clojure.core/load-lib
                  RestFn.java:142 clojure.lang.RestFn.applyTo
                     core.clj:648 clojure.core/apply
                    core.clj:5774 clojure.core/load-libs
                    core.clj:5758 clojure.core/load-libs
                  RestFn.java:137 clojure.lang.RestFn.applyTo
                     core.clj:648 clojure.core/apply
                    core.clj:5796 clojure.core/require
                    core.clj:5796 clojure.core/require
                  RestFn.java:408 clojure.lang.RestFn.invoke
                     eval.clj:105 lighttable.nrepl.eval/require-BAR-create-ns
                     eval.clj:103 lighttable.nrepl.eval/require-BAR-create-ns
                     eval.clj:173 lighttable.nrepl.eval/eval2472[fn]
                 MultiFn.java:229 clojure.lang.MultiFn.invoke
                      core.clj:97 lighttable.nrepl.core/queued[fn]
                    core.clj:2459 clojure.core/comp[fn]
       interruptible_eval.clj:159 clojure.tools.nrepl.middleware.interruptible-eval/run-next[fn]
                      AFn.java:22 clojure.lang.AFn.run
     ThreadPoolExecutor.java:1142 java.util.concurrent.ThreadPoolExecutor.runWorker
      ThreadPoolExecutor.java:617 java.util.concurrent.ThreadPoolExecutor$Worker.run
                  Thread.java:748 java.lang.Thread.run

It seems there is an issue with the html address being used for tensorboard projection. Could anyone tell me what is the issue here exactly and what can I do about it?

[Harold]

Hi,

That error indicates that the port that the classification example is trying to use to display its web page is already in use. In general, only one application can use a port at any given time.

Typically this happens if you try to run two copies of the example at the same time (on purpose or on accident).

Two things I'd try:

1) Kill any processes that could be running a server on that port, especially other java processes running the classification example code.

2) Change the port that the example is using. This is the relevant line of code:

 - https://github.com/thinktopic/cortex/blob/master/experiment/src/cortex/experiment/classification.clj#L225

Hope the helps. Let us know if there's more we can do,

-Harold