Jayendra

How about using [custom metrics](https://cloud.google.com/monitoring/api/metrics_gcp), and having [monitoring dashboard](https://cloud.google.com/monitoring/dashboards) ? This requires modification in the existing pipeline code, where it updates the our custom metrics, and dashboard will be able...

yes, this is indeed a very useful feature, I can contribute to add it if someone points me to that. Is there any quick and dirty way to import the...

I am also getting the same error, where I am trying to train the standard TrnasformerLM, with the slight modification in the parameters(vocab_size=50000, max_len=1024). The same code works on the...

This `stream` msg would help to identify the error, note the `stderr` inside the `content`. ``` {'buffers': [], 'content': {u'name': u'stderr', u'text': u'\x1b[0;31mError: This statement is incomplete.\n\n\x1b[0m'}, 'header': {u'date': datetime.datetime(2018,...

yes output from `stderr` doesn't always indicate a error, for example `fprintf(2,'Hello')` would write to `stderr`, and it is not erroneous code. We can't use output from `stderr` as flag...

Can you help me on debugging that ? Currently I am trying this, but I don't think this is the way to run the kernel. ``` In [1]: from matlab_kernel.kernel...

same things is happening with python2 and python3 ``` Python 3.6.6 (default, Sep 12 2018, 18:26:19) [GCC 8.0.1 20180414 (experimental) [trunk revision 259383]] on linux Type "help", "copyright", "credits" or...

I have checked the https://github.com/Calysto/metakernel/issues/149 which is similar to this scenario, except the octave vs matlab distinction. With matlab.engine I think this should be easier to do for matalab case...