DL solution without any decisions regarding the neural architecture of the model

microsoft / nni

An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.

https://nni.readthedocs.io

MIT License

13.99k stars 1.81k forks source link

DL solution without any decisions regarding the neural architecture of the model #3830

Open kvartet opened 3 years ago

kvartet commented 3 years ago

Describe the issue:

does NNI have a solution for DL that allows me to feed in the inputs, outputs and evaluate the model while leaving me out of any decisions regarding the neural architecture of the model? Most of the examples I have seen force me to make some layer choices

Originally posted by @8W9aG in Gitter

ultmaster commented 3 years ago

Good point. We will design and try to support this feature in future releases.

8W9aG commented 3 years ago

Thanks for looking into this, some features that would be great are:

Feed in a current model that works, and maybe start the search from
Search is never-ending (ie it keeps going further and further from the current model)

Is there anything I can do to help this along?