sampling based on divergence
Project description
AutoRA Divergence Experimentalist
The divergence experimentalist identifies experimental conditions $\vec{x}' \in X'$ with respect the distance between existing experimental data $\vec{x}, $\vec{y} and data predicted by a model $\vec{x_pool}, $\vec{y_pred}:
$$ \underset{\vec{x}'}{\arg\max}~sum(d((\vec{x}, \vec{y}), (\vec{x_pool}, \vec{y_pred})) $$
The aim of this experimentalist is to combine novelty and uncertainty by using a distance that combines both: The distance between existing conditions to new conditions and the distance of existing observations to predictions.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for autora_experimentalist_divergence-0.0.2.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | a54f5652acffc866f5ff0945fe0faa22456d1ceba3f1cf7ae7142fbcb2d02620 |
|
MD5 | 15ff845ea73c986e1a6c26ced9631f6f |
|
BLAKE2b-256 | 19c05d06458d879472d2b9eaafc863b742f8747d340c7eb27e05e785d428a28a |
Close
Hashes for autora_experimentalist_divergence-0.0.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3158a0bd09dab7541a1b202050810460147990472c8eca327a7e58c1d86fa556 |
|
MD5 | 44b5c0fd66f7ac6af188034be1a2e10e |
|
BLAKE2b-256 | c572a6a26feb3f97ec6f955d788c506bf53202c0236f98752ea8ce10a9edf042 |