Snapshots: ML scripts for reproducibility
This is a set of scripts to generate results of our paper: Statistical Analysis and Discovery of Heterogeneous Catalysts Based on Machine Learning from Diverse Published Data (K. Suzuki, T. Toyao, Z. Maeno, S. Takakusagi, K. Shimizu, I. Takigawa)
The results are based on the following versions of related libraries. To reproduce the same outputs, you need these versions on anaconda3-5.0.1.
- sklearn: 0.19.1
- xgboost: 0.81
- pandas: 0.23.1
- numpy: 1.14.5
- scipy: 1.0.1
- matplotlib: 2.2.4
- Keisuke Suzuki (Hokkaido University)
- Ichigaku Takigawa (https://itakigawa.github.io/)