DeepNano with Metrichor: Bad input argument to theano function

[Natasha Pavlovikj]

When I try to run DeepNano with Metrichor, I get the following error: TypeError: ('Bad input argument to theano function with name "rnn_fin.py:81" at index 0(0-based)', 'Wrong number of dimensions: expected 2, got 1 with shape (0,).')

I can see this is fixed for DeepNano without Metrichor (Issue #3), so I would appreciate if you can fix it in basecall.py as well.

[Natasha Pavlovikj]

I realized that the error occurred because I was using reads generated with SQK-NSK007 chemistry, and not SQK-NSK005 or SQK-NSK006. Do you think soon you can support reads generated with SQK-NSK007 chemistry ?

[Vlado Boza]

We are preparing training data for R9.4 soon. I expect everything to be finalized in two to three weeks.

[Natasha Pavlovikj]

Thank you so mcuh @vboza , I am looking forward to the new release soon :)