rdisipio / DeepBayesIris

Iris dataset separation using Bayesian Neural Network

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DeepBayesIris

Requirements:

ROOT: https://root.cern.ch/

BAT: https://github.com/bat/bat

Installation

Install ROOT

Downloading binaries is recommended over compiling source files.

To set it up:

source ${HOME}/local/root/bin/thisroot.sh

Install BAT

./configure --prefix=$HOME/local
make
make install

Make sure the BAT libraries are available (follow instructions to set up environment variables)

Compile DeepBayesIris

git clone https://github.com/rdisipio/DeepBayesIris.git
cd DeepBayesIris
make

Training

./runDeepBayesIris

Inference

Run:

./predict [eventId] [numberOfRuns]

Inspect output file:

root predict_0.root
root [0] 
Attaching file predict_0.root as _file0...
(TFile *) 0x7faac86e2830
root [0] output->Draw("lego")

Make plots

To make nice-looking plots:

root
root [0] .L make_plot.C 
root [1] make_plot(1) // event id=1
root [2] make_plot(70) // event id=70

About

Iris dataset separation using Bayesian Neural Network


Languages

Language:C++ 94.0%Language:C 3.1%Language:Makefile 2.9%