Passing multiple logdirs doesn't work (and raises warning about "Conflict for name")

These work:

tensorboard --logdir logs/foo
tensorboard --logdir logs/bar
tensorboard --logdir a:logs/foo,b:logs/bar

But when I run this:

tensorboard --logdir logs/foo,logs/bar

I get a warning like this repeatedly printed:

WARNING:tensorflow:Conflict for name .: old path /home/colin/src/myproject/logs/foo, new path /home/colin/src/myproject/logs/bar

In tensorboard, only one run is shown.

I installed tensorflow-tensorboard using pip, and the version is 0.1.2

(I tried debugging, but the warning gets raised in a separate daemon thread, and I wasn’t able to figure out how to get pdb to attach to it)

[wchargin]

@colinmorris you can attach pdb as follows:

diff --git a/tensorboard/backend/event_processing/event_multiplexer.py b/tensorboard/backend/event_processing/event_multiplexer.py
index 8a71f82..66e576a 100644
--- a/tensorboard/backend/event_processing/event_multiplexer.py
+++ b/tensorboard/backend/event_processing/event_multiplexer.py
@@ -124,6 +124,7 @@ class EventMultiplexer(object):
         if name in self._paths and self._paths[name] != path:
           # TODO(danmane) - Make it impossible to overwrite an old path with
           # a new path (just give the new path a distinct name)
+          import pdb; pdb.set_trace()  # XXX BREAKPOINT
           tf.logging.warning('Conflict for name %s: old path %s, new path %s',
                              name, self._paths[name], path)
         tf.logging.info('Constructing EventAccumulator for %s', path)

(You must invoke with bazel-bin/tensorboard/tensorboard --logdir /tmp/a,/tmp/b rather than bazel run tensorboard -- --logdir /tmp/a,/tmp/b.)

[shwang]

Here’s a simple way to make this work that avoids the overlapping run names issue.

We could add another flag --log_dir_multi [log_dir1] [log_dir2] ....

It errors if the --log_dir option is also present. To process this flag, Tensorboard first makes a temporary directory $gather_dir. Then Tensorboard makes a symlinks all the log dirs in $gather_dir/[log_dir#]. Finally Tensorboard proceeds as if we ran tensorboard --log_dir $gather_dir.

The run names will look as if you ran tensorboard --log_dir . but only with the selected log dirs.