Compares weaknesses in multiple models
Project description
AutoRA Model Disagreement Experimentlist
The model disagreement experimentlist identifies experimental conditions $\vec{x}' \in X'$ with respect to a pairwise distance metric between theorist models, $P_{M_{i}}(\hat{y}, \vec{x}')$:
$$ \underset{\vec{x}'}{\arg\max}~(P_{M_{1}}(\hat{y}, \vec{x}') - P_{M_{2}}(\hat{y}, \vec{x}'))^2 $$
Example Code
from autora.experimentalist.model_disagreement import model_disagreement_sample
from autora.theorist.bms import BMSRegressor; BMSRegressor()
from autora.theorist.darts import DARTSRegressor; DARTSRegressor()
import numpy as np
#Meta-Setup
X = np.linspace(start=-3, stop=6, num=10).reshape(-1, 1)
y = (X**2).reshape(-1, 1)
n = 5
#Theorists
bms_theorist = BMSRegressor()
darts_theorist = DARTSRegressor()
bms_theorist.fit(X,y)
darts_theorist.fit(X,y)
#experimentlist
X_new = model_disagreement_sample(X, [bms_theorist, darts_theorist], n)
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_model_disagreement-2.1.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5b26bad6e6596bd313cf59f7c762ac2fd04c06a45f81e88d8b52802edfb6f88f |
|
MD5 | 86c6aeeeceb19a5bef0804f8077725c2 |
|
BLAKE2b-256 | d476dc2d745884aecb276815a4d3064aad0badb2e6709cbcd8e66e5cdac665c5 |
Close
Hashes for autora_experimentalist_model_disagreement-2.1.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | f7c87e8da60b84d50a24d21f65733b7bd6d5b28f1335fc65014af516a54db79b |
|
MD5 | 69629c1bbeb08f355a38086427438448 |
|
BLAKE2b-256 | b3b2ca0ee7a5ca3113e48924d90b1d9c1929cdc27780af5adbcb8ba16815fab1 |