For DeepNano without Metrichor

[Ju Qiu]

When read the raw fast5 files, the error messages caused the program stop. But after I removed the problem read, the program run until the next error read. How to filter out the problem reads and save time on running the program?

TypeError: ('Bad input argument to theano function with name "/deepnano/rnn_ fin.py:83" at index 0(0-based)', 'Wrong number of dimensions: expected 2, got 1 with shape (0 ,).')

[Vlado Boza]

I added option for this.

[TCherici]

May I know what the option is to solve this problem?

[Natasha Pavlovikj]

@TCherici , I guess that would be the "--max_2d_length" option for DeepNano without Metrichor.