MXBoard with keras-mxnet

[sdbonte]

• [x] Check that you are up-to-date with the master branch of Keras. You can update with: pip install git+git://github.com/keras-team/keras.git --upgrade --no-deps

• [x] Check that your version of TensorFlow is up-to-date. The installation instructions can be found here.

• [x] Provide a link to a GitHub Gist of a Python script that can reproduce your issue (or just copy the script here if it is short).

Is MXBoard supported in Keras with MXNet backend? I managed to get my network (which I previously used with the TensorFlow backend) training with the MXNet backend, but now I want to monitor the performance as I was used to with TensorBoard. This is what I tried:

from keras.losses import mean_squared_error
sw = SummaryWriter(logdir="logs/test", flush_secs=10)
sw.add_scalar(tag='training_MSE', value=mean_squared_error)

but this gives the following error: TypeError: _make_numpy_array only accepts input types of numpy.ndarray, scalar, and MXNet NDArray, while received type <class 'function'>

How can MXBoard be used in Keras?

[sdbonte]

Solved: Keras allows to use the TensorBoard module in combination with MXNet backend.