You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Dragonn (http://kundajelab.github.io/dragonn/) is the most mature deep learning library for genomic data. Would it be possible to adapt dragonn models for drug discovery?
The text was updated successfully, but these errors were encountered:
As far as as I recall, DeepChem still doesn't have full functionality of dragonn ported in right?
There are also some custom layers, a few of which I found in Avanti's paper and few from the Janggu library, which could be ported into DeepGene pr dc.genomics.
With TF 2.0 based on tf.keras and tf.contrib going away, it might be easier to integrate DragoNN now.
Relevant discussion #1461.
Yes, agreed. We're going to start pushing to integrate DeepChem layers with tf.keras layers so we get full TF 2.0 inter-op. This will be the next big push after DeepChem 2.2 is cut.
Dragonn (http://kundajelab.github.io/dragonn/) is the most mature deep learning library for genomic data. Would it be possible to adapt dragonn models for drug discovery?
The text was updated successfully, but these errors were encountered: