Machine Learning the Voltage of Electrode Materials in Metal-Ion Batteries

This repository is the reproduction for "Machine Learning the Voltage of Electrode Materials in Metal-Ion Batteries".

Repository

├── dataset
├── preprocess              # feature engineering scripts
├── regression              # train scripts
├── .flake8
├── .gitignore
├── README.md
└── requirements.txt

How to reproduce

Python version : 3.6.5

1. Clone this repository and setup the environment

$ git clone https://github.com/nissy-dev/machine-learning-the-voltage-of-electrode-materials.git
$ cd machine-learning-the-voltage-of-electrode-materials
$ python -v venv venv
$ source venv/bin/activate
$ pip install -r requirements.txt

2. Go to the preprocess folder and create features

$ cd preprocess
$ python create_features.py

3. Go to the regression folder and train models

YYYY_MM_DD is the date you created features. Please set your date.

$ cd regression
$ python train.py --feat-path ../dataset/feature/repro_features_YYYY_MM_DD.csv --method svr --out-dir result/svr
$ python train.py --feat-path ../dataset/feature/repro_features_YYYY_MM_DD.csv --method krr --out-dir result/krr
$ python train_dnn.py --feat-path ../dataset/feature/repro_features_YYYY_MM_DD.csv --out-dir result/dnn

Result

	SVR	KRR	DNN	*SVR	*KRR	*DNN
fold 1	0.41	0.45	0.50	0.51	0.54	0.42
fold 2	0.43	0.44	0.52	0.25	0.28	0.48
fold 3	0.45	0.45	0.47	0.26	0.27	0.42
fold 4	0.38	0.41	0.46	0.35	0.47	0.44
fold 5	0.42	0.42	0.46	0.38	0.43	0.44
fold 6	0.40	0.41	0.46	0.62	0.71	0.42
fold 7	0.40	0.41	0.44	0.43	0.42	0.43
fold 8	0.42	0.42	0.46	0.59	0.62	0.42
fold 9	0.40	0.46	0.47	0.53	0.57	0.45
fold 10	0.41	0.42	0.51	0.28	0.30	0.48
10 fold MAE (mean±std)	0.42±0.02	0.43±0.02	0.48±0.03	0.42±0.13	0.46±0.14	0.43±0.03
H-test	0.45	0.47	0.54	0.40	0.39	0.43
Na-test	(0.63)	(0.59)	(0.61)	1.00	0.93	1.25

*SVR, *KRR, *DNN are reported in the paper.

References

[1] Rajendra P. JoshiJesse, et al. Machine Learning the Voltage of Electrode Materials in Metal-Ion Batteries. ACS Appl. Mater. Interfaces, 18494-18503, 2019.
[2] Anubhav Jain, Shyue Ping Ong, Geoffroy Hautier, Wei Chen, William Davidson Richards, Stephen Dacek, Shreyas Cholia, Dan Gunter, David Skinner, Gerbrand Ceder, et al. Commentary: The materials project: A materials genome approach to accelerating materials innovation. Apl Materials, 1(1):011002, 2013