search

nanoporetech / bonito

A PyTorch Basecaller for Oxford Nanopore Reads
https://nanoporetech.com/
Other
389 stars 120 forks source link

Tweaking model parameters #47

Closed cnk113 closed 4 years ago

cnk113 commented 4 years ago

Hello,

I'm trying to test possible enhancements based on ContextNet, but I've noticed the current model file has now deviated from Quartznet. I was wondering how you were able to arrive on the architecture? Was it using some type of grid search? Since ContextNet uses similar architecture compared to Quartznet in the encoder layer I was considering using similar parameters from Bonito.

iiSeymour commented 4 years ago

Hey @cnk113

That sounds like a great idea.

The deviated model was arrived at by doing a large model sweep using bonito tune that uses optuna.

See - https://github.com/nanoporetech/bonito/blob/master/bonito/tune.py