ThompsonA93 / CIFAR10_Benchmark

Neural Network experimentation on the CIFAR-10 dataset ( https://www.cs.toronto.edu/~kriz/cifar.html )

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

CIFAR10_Benchmark

Design

Environment Version
Operating System Ubuntu 20.04.4 LTS, Windows 10 21H2
Python 3.8.10
PIP pip 22.1.2
Experimental Setup Component
Windows 10 21H2 CPU: 11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz 1.80 GHz
RAM: 32.0 GB (31.4 GB usable)
Ubuntu 20.04.5 LTS CPU: Intel® Core™ i7-8700 CPU @ 3.20GHz × 12
RAM: 31,2 GiB

System

Installation

The installation of python depends per operating system.

The python dependencies can be installed using REQUIREMENTS.txt (see INSTALL.sh).

Execution

There are two distinct ways to run code located in ''/src'':

  1. Jupyter: The IPYNB files are considered the main files. It is recommended to use the program Visual Studio Code with the Jupyter-Extension. Any alternative may also work.
  2. Native python: The PY files are simply copied from the content of the Jupyter files. Mainly used in context of the RUN scripts (Powershell or Shell) for multiple execution.

Note that ''/src/config.py'' contains additional settings which are relevant for both, IPYNB and PY files.

The execution of either Jupyter or native python files will create log-files in ''/log'', which contain the most crucial information on the programs performance.

Troubleshooting / Errors

None as of now.

Documentation

Report

Available in tex

Log-Files

Some are archived in log-lnx and log-win. New logs are generated as the python scripts are executed.

About

Neural Network experimentation on the CIFAR-10 dataset ( https://www.cs.toronto.edu/~kriz/cifar.html )


Languages

Language:Jupyter Notebook 52.2%Language:Python 31.7%Language:TeX 12.4%Language:R 3.3%Language:Shell 0.2%Language:PowerShell 0.2%