Skip to content

Evaluation of Classifier Algorithms for Polyolefin Sorting using NIR Spectroscopy

License

Notifications You must be signed in to change notification settings

usnistgov/ECAPS

Repository files navigation

Evaluation of Classifier Algorithms for Polyolefin Sorting using NIR Spectroscopy

This repository supports the following submitted manuscript:

Bradley P. Sutliff, Peter A. Beaucage, Debra J. Audus, Sara V. Orski, and Tyler B. Martin, "Sorting Polyolefins with Near-visible Infrared Spectroscopy: Identification of optimal data analysis pipelines and machine learning classifiers" Digital Discovery 2024 NUM (NUM), PAGES DOI:

which compares near-visible infrared spectroscopy coupled with various data preprocessing steps and machine learning techniques to classify polyolefin subclasses and enable better sorting of polyolefins. The repository is intended for the following use cases:

  • Illustrate key ideas from the submitted manuscript including using various preprocessing strategies that are commonly applied to near-visible infrared spectra.
  • Reproduce figures from our submitted manuscript.
  • Allow for full reproducibility of the data in the submitted manuscript.

Running the notebook

The included notebooks allow the user to reproduce all of the figures in the corresponding submitted manuscript, using the corresponding data. The notebooks are designed to sequentially walk the user through each of the prepossessing and analyzing steps, generating corresponding data checkpoints and figures along the way. All code is written in Python and requires Python == 3.9. It can be used on any operating system. Other requirements are listed in requirements.txt.

The code can either be run locally or in Google Colab.

Run notebook locally (option 1)

Clone the code

First clone the code via

git clone https://github.com/usnistgov/ecaps.git

and navigate to the directory where the repository lives

cd ecaps

Next, one needs to create a virtual environment. This can be done using Python virtual environments or with Anaconda. Both options are listed below.

Create a Python virtual environment (option 1)

First, make sure you are using Python 3.9.

python3 -m venv env

where env is the location of the virtual environment

Activate the virtual environment

source env/bin/activate

Install dependencies

python3 -m pip install -r requirements.txt

Create a virtual environment with Anaconda (option 2)

First, install conda. Then run the following in a conda-enabled terminal.

conda env create -f ecaps.yml

This yml file will automatically name the new environment ecaps.

If you are using conda>=4.6, activate the virtual environment via

conda activate ecaps

Otherwise, see the conda docs

Run the notebook

The Jupyter notebook can by run by using the command

jupyter notebook

Running notebook in Google Colab (option 2)

If you are interested in running the notebook in Google Colab, first click on the relevant link below. Note that these links were generated by navigating to the notebook of interest on the nir_corr_po GitHub page, for example, https://github.com/usnistgov/pirc/1-pirc_PreProcessingExamples.ipynb and then replace github.com with githubtocolab.com.

This should open the notebook in Google Colab.

Contact

Bradley P. Sutliff, PhD
Materials Science and Engineering Division
Material Measurement Laboratory
National Institute of Standards and Technology

Email: Bradley.Sutliff@nist.gov
GithubID: @bpsut
Staff website: https://www.nist.gov/people/bradley-sutliff

How to cite

If you use the code, please cite our submitted manuscript:

Bradley P. Sutliff, Peter A. Beaucage, Debra J. Audus, Sara V. Orski, and Tyler B. Martin, "Sorting Polyolefins with Near-visible Infrared Spectroscopy: Identification of optimal data analysis pipelines and machine learning classifiers" Digital Discovery 2024 NUM (NUM), DOI:

If you use the data, please cite:

Sutliff, Bradley P., Goyal, Shailja, Martin, Tyler B., Beaucage, Peter A., Audus, Debra J., Orski, Sara V. (2023), Correlating Near-Infrared Spectra to Bulk Properties in Polyolefins, National Institute of Standards and Technology, https://doi.org/10.18434/mds2-3022 (Accessed 2024-07-03)

Future updates and maintenance

There is no intent to update/maintain this repository once it is released to the public, given that it is intended to reproduce published figures and analysis that should not change over time.

About

Evaluation of Classifier Algorithms for Polyolefin Sorting using NIR Spectroscopy

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages