This repository contains a Python implementation of the AdaptFTIR method described in the paper entitled "Title" (doi url).
The package is available via PyPI:
pip install AdaptFTIRFor detailed information on how to initialize the AdaptFTIR method and configure its parameters, please refer to the example_usage.ipynb Jupyter Notebook and the code documentation in the AdaptFTIR.py file. The basic usage format is as follows:
from AdaptFTIR import AdaptFTIR
# Initialize an instance of AdaptFTIR with a calibration set
adapter = AdaptFTIR(spectra=<your_spectra_to_augment>,
subject_ids=<your_subject_ids_for_spectra_to_augment>,
calibration_set=<your_calibration_set>,
calibration_ids=<your_calibration_set_subject_ids>)
# Create synthetic samples based on an input set of training samples X and associated sample labels y
X_gen, y_gen = AdaptFTIR.run(n_per_subject_id=<your_n_per_subject_id>)Citation goes here