hakula139/resource_usage_prediction

A multi-threaded time-series-based real-time server resource usage predictor, trained dynamically on a GRU model, developed with LibTorch (C++)

resource_usage_prediction

A resource usage predictor, developed with PyTorch (C++ frontend), written in modern C++.

Table of Contents

• resource_usage_prediction
  • Table of Contents
  • Getting Started
    • Prerequisites
    • Installation
    • Usage

Getting Started

Prerequisites

Before installation, you need to have the following dependencies installed.

• CMake 3.5 or later
• Gnuplot

Installation

Execute ./scripts/build.sh to build the project using CMake.

Usage

Execute ./bin/generator to generate a testing dataset.

Execute ./bin/predictor to start predicting on the testing dataset.