This repository shows the code used in the publication "Inception for Petroleum Analysis" for cetane number prediction.
The preprint is currently under peer evaluation at Fuel (Elsevier).
Information : y is already normalized (mean 0 & std 1), the scaler is not given. Therefore, calculated metrics will be 15 times smaller than on the publication.
Follow the next commands:
conda create -n inception-petroleum-analysis python=3.10
conda activate inception-petroleum-analysis
conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0
pip install "tensorflow<2.11"
pip install -r requirements.txt
Install MATLAB R2022b and Eigenvector PLS_Toolbox version 9.2.
Open the script within the eponymous software.
Activate the environment, then
python -m scripts.IPA-calibration
python -m scripts.DeepSpectra-calibration