I created Tensorboard logs with a graph, using the tutorials/mnist/fully_connected_feed.py script (I also have the same issue on other scripts).
summary_writer = tf.train.SummaryWriter("logdir", graph_def=sess.graph_def)
The Tensorboard works well for Scalars, Images and Histograms, but not for graphs. I get a white page, with only the header.
If I open the browser console (Firefox 42, Ubuntu 14.04), I get those errors when I switch to the GRAPH tab:
NS_ERROR_FAILURE: tf-tensorboard.html-12.js:115:0
Reading graph.pbtxt : 2 ms tf-tensorboard.html.js:13:1
Reading stats.pbtxt : 0 ms tf-tensorboard.html.js:13:1
Parsing graph.pbtxt : 26 ms tf-tensorboard.html.js:13:1
Normalizing names : 10 ms tf-tensorboard.html.js:13:1
Building the data structure : 13 ms tf-tensorboard.html.js:13:1
Adding nodes : 14 ms tf-tensorboard.html.js:13:1
Detect series : 16 ms tf-tensorboard.html.js:13:1
Adding edges : 14 ms tf-tensorboard.html.js:13:1
Finding similar subgraphs : 27 ms tf-tensorboard.html.js:13:1
new tf.graph.render.Hierarchy : 17 ms tf-tensorboard.html.js:13:1
tf-graph-scene (layout): : 214 ms tf-tensorboard.html.js:13:1
tf-graph-scene (build scene): : 118 ms tf-tensorboard.html.js:13:1
NS_ERROR_FAILURE: tf-tensorboard.html-12.js:115:0
NS_ERROR_FAILURE: tf-tensorboard.html-6.js:76:0
TypeError: this.translate is undefined
How to fix this ?
Tensorboard is only working in Google Chrome as far as I can tell. I tried Safari, Firefox and Chrome on my Mac and it only showed the graph properly in Chrome.