david-thrower / cerebros-core-algorithm-alpha

The Cerebros package is an ultra-precise Neural Architecture Search (NAS) / AutoML that is intended to much more closely mimic biological neurons than conventional neural network architecture strategies.
Other
27 stars 4 forks source link

lightweight-testing-on-tendem-embeddings #135

Open david-thrower opened 11 months ago

david-thrower commented 11 months ago

Kind of issue: enhancement

Expected behavior A clear description of what you expected to happen.

Additional context: Alex suggested to try injecting embedding layers into the NAS. Noting that embedding layers only take discrete int inputs, I added that bucketization of the output of a Dense() or Concatenate() layer before the embedding is needed to make this work.

Here we can try this approach.

Suggested Labels (If you don't know, that's ok): kind/enhancement kind/performance