Tzu-Wei Huang

Hi, I believe you are saying that show the AUC value in the tensorboard PR_curve page. This feature request should belong to tensorflow/tensorboard since it's frontend related.

I can't determine from that. Which program spit that message? Did that stop your training?

A reproducible code would be great help! thanks There is a `writer.close()` method. Did you close the old writer before opening a new one?

Hi, what is your protobuf version?

@TengdaHan Can you provide more info?

@jendrikjoe Thanks for the investigation and nice spot. I am curious if using the default `tensorflow` binning solves your problem too. If you would like to help to do the...

Can you share your model? Is there any custom layer used?

I think it's doable. Looks like the data is pickled during training and unpickled with the webserver. ``` W0530 23:39:30.068949 Thread-4 application.py:274] path /[[_dataImageSrc]] not found, sending 404 W0530 23:39:30.076314...

@marcociccone I partially implemented that before and here's the hidden demo: (I think that is fast enough, please compare it with yours.) https://github.com/lanpa/tensorboardX/blob/master/examples/demo_beholder.py and the code:(which is still dirty:) https://github.com/lanpa/tensorboardX/tree/master/tensorboardX/beholder

Thank you for the bug report. I have tried your example but find that I have an opposite result compare to yours. (Example 1 has four columns, Example 2 has...