cerebros-core-algorithm-alpha
cerebros-core-algorithm-alpha copied to clipboard
david-thrower
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 strategi...
A conv1d example, for example time series, like sames forecasting: This can be used as a template for the embedding to duplicate for this: https://colab.research.google.com/github/kmkarakaya/ML_tutorials/blob/master/Conv1d_Predict_house_prices.ipynb#scrollTo=A4JNcPxXqrBo This may be a suitable...
1. All param for list of activations 2. Parameter for probably of a given option [e.g. list of n activations to be tried as one param, list of n probabilities,...
Required for github community guidelines ratings ...
If we add a word2vec or tfidf input embedding to a Cerebros module and make 500 separate softmax outputs, 1 for each of 500 sequential words, then an argmax then...
Add an ancillary module or class that loads data like the notebook and searches a problem - specific presumed feasible space. Dependencies: Fix issues with race conditions with past attempts...
Metadata
Owner
Metadata
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 strategi...