Scripts for machine learning at ONR project (2017-)
Prerequisites: Python 2, Numpy, Sklearn and Matplotlib Packages
Data: unnorma.input.data contains four columns, which are Temperature, Concentration, Dwelling Time and %Mass Change. norm.input.data contains four columns after normalization, consistent with unnorma.input.data.
Machine Learning: Four machine learning models are explored based on 19 bulk sample results. The target property is %Mass Change and they're normalized before the machine learning process.
- Multivariate Linear Regression
http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html
- Support Vector Regression
http://scikit-learn.org/stable/modules/generated/sklearn.svm.SVR.html
- Kernel Ridge Regression
http://scikit-learn.org/stable/modules/generated/sklearn.kernel_ridge.KernelRidge.html
- Neural Network