README.md

dhsvm-opt - Optimization library for DHSVM

Table of Contents

1. Introduction
2. Directory Structure
3. References
4. Setup

Introduction

dhsvm-opt is an end to end solution to tune and optimize DHSVM physical parameters using Markov Chain Monte Carlo simulation algorithm DFREAM implemented in spotpy and using Machine learning methods like Boosted Trees.

Directory Structure

.
├── data
├── scripts
    ├── fast
    ├── dream
    ├── decision_tree
    ├── sql-setup

Setup

Run setup file to get the custom sql version installed to handle 30,000 columns. Running below sets of command would setup a conda environment dhsvm-opt with custom sqlite version.

bash -i setup.sh
source ~/.bashrc
conda activate dhsvm-opt

References

Literature

1. DREAM -an adaptive Markov Chain Monte Carlo simulation algorithm to solve discrete, noncontinuous, and combinatorial posterior parameter estimation problems

Implementations

1. spotpy- A Statistical Parameter Optimization Tool