search

intel-analytics / analytics-zoo

Distributed Tensorflow, Keras and PyTorch on Apache Spark/Flink & Ray
https://analytics-zoo.readthedocs.io/
Apache License 2.0
18 stars 4 forks source link

Tensorboard support for Jupyter notebook server. #1070

Open hhbyyh opened 5 years ago

hhbyyh commented 5 years ago

User has a Spark cluster and also a jupyter notebook server. They can access the jupyter notebook server from many clients.

Now if they want to use the training summary and validation summary, is there a way to put all the logs to the client node, rather than the jupyter server node. I know users can use hdfs to store the training log, but it will be hard to monitor the training progress at run time.

hhbyyh commented 5 years ago

Ah Tensorboard may support hdfs path by now. I'm running some test.

hhbyyh commented 5 years ago

It does work. But user need to specify some env variables.

https://github.com/tensorflow/examples/blob/master/community/en/docs/deploy/hadoop.md

hhbyyh commented 5 years ago

Closing the issue now.

jason-dai commented 5 years ago

We need to add this to our documentation for visualization @qiuxin2012