CrossTrainer: Practical Domain Adaptation with Loss Reweighting
Project description
CrossTrainer: Practical Domain Adaptation with Loss Reweighting
This is an implementation of the method described in "CrossTrainer: Practical Domain Adaptation with Loss Reweighting" by Justin Chen, Edward Gan, Kexin Rong, Sahaana Suri, and Peter Bailis.
Install
The crosstrainer package can be installed using pip.
pip install crosstrainer
Usage
CrossTrainer utilizes loss reweighting to train machine learning models using data from a target task with supplementary source data.
Inputs:
Base model, target data, source data.
Outputs:
Trained model with optimized weighting parameter alpha.
Example:
import crosstrainer
from sklearn import linear_model
lr = linear_model.LogisticRegression()
ct = CrossTrainer(lr, k=5, delta=0.01)
lr, alpha = ct.fit(X_target, y_target, X_source, y_source)
y_pred = lr.predict(X_test)
More examples can be found in the tests file: crosstrainer/tests/test_crosstrainer.py
.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
No source distribution files available for this release.See tutorial on generating distribution archives.
Built Distribution
Close
Hashes for crosstrainer-0.1.5-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d0754f4a5e47ada8721d9a4977e551383a4ed318f8cd22327fc5b78be7eef049 |
|
MD5 | 4b03a25243c91b5e4868988bd47ad906 |
|
BLAKE2b-256 | aa13a3435c3873b048c31588a4ca3fc9edfe9aa7ccf1011491835031ae23e3a0 |