Tensorboard gives empty page

2,615 views
Skip to first unread message

yitzhak....@gmail.com

unread,
Jun 25, 2017, 5:38:27 AM6/25/17
to Discuss
Hi, trying to visualize stuff with tensorboard (I am using keras) the browser doesn't show anything. Here are the relevant lines from my Train.py

tbCallBack = keras.callbacks.TensorBoard(log_dir='./Graph', histogram_freq=0, write_graph=True, write_images=True)
train_hist
=model.fit(x_train, y_train,
         
batch_size=batch_size,
         
epochs=epochs,
         
validation_data=(x_test, y_test),
         
callbacks=[tbCallBack])

Now the training itself works fine. it shows the iterations trough the epochs etc. Now while the Train.py is running I run
$ tensorboard --logdir=./Graph
I see some event file being created in the ./Graph directory and I tensorboard gives me a link for the browser. However when I open it, it is empty. I tried it both in Chrome and in Firefox. 

Krishnan Srinivasan

unread,
Jun 27, 2017, 2:03:57 AM6/27/17
to Discuss, yitzhak....@gmail.com
I was running to this issue recently and found that it had to do with allowing Chrome to load javascript/images/popups for tensorboard (click on the information icon next to the url). Haven't narrowed down the problem any further however. Additionally, if you have another browser like Safari, try to open tensorboard on there and see if it works (it did for me).

yitzhak....@gmail.com

unread,
Jun 27, 2017, 2:15:23 AM6/27/17
to Discuss, yitzhak....@gmail.com
the issue is solved now. here it is:

Sebastian Raschka

unread,
Jun 27, 2017, 2:41:59 AM6/27/17
to TensorFlow Mailinglist
> However when I open it, it is empty. I tried it both in Chrome and in Firefox.


Not sure if this is necessary (or already done) via Keras, but when I use tensorboard with tensorflow, I sometimes need to have a manual "flush" the writer to see the results

e.g., as in

with tf.Session(graph=g) as sess:

sess.run(tf.global_variables_initializer())

file_writer = tf.summary.FileWriter(logdir='logs/1', graph=g)

for i in range(5):
result, summary = sess.run([update_tf_x, merged_summary],
feed_dict={some_value: i})

file_writer.add_summary(summary=summary, global_step=i)
file_writer.flush()


Just an idea that this could maybe be an issue here?
> --
> You received this message because you are subscribed to the Google Groups "Discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to discuss+u...@tensorflow.org.
> To post to this group, send email to dis...@tensorflow.org.
> To view this discussion on the web visit https://groups.google.com/a/tensorflow.org/d/msgid/discuss/10ddf35f-ea60-4e7f-bf69-d44d0954a054%40tensorflow.org.

Reply all
Reply to author
Forward
0 new messages