HandWritting-Digit-Detection

Detect handwritten digit from images using scikit-learn. This is a final project for ET3107 "Pemrograman Lanjut" Course.

Installation

NOTES: This program currently only support Ubuntu. For windows, installation can be done using wsl. Clone this repository

git clone https://github.com/AdityaP1502/HandWritting-Digit-Detection
cd HandWritting-Digit-Detection

First, install virtual environment and create a new virtual environment on python

python3 -m pip install --user virtualenv &&
sudo apt install python3-venv &&
python3 -m venv env

then, activate the virtual environment by running

source env/bin/activate

then, grant permission to install.sh scripts by running command below

chmod +x ./scripts/install.sh

then, to install just run the scripts

./scripts/install.sh

Notes:

For windows user running in wsl, if found error

./scripts/build.sh: line 3: $'\r': command not found
./scripts/build.sh: line 69: syntax error: unexpected end of file

Change the EOL from CRLF(windows) TO LF(LINUX). You can use notepad++ EOL conversion for this. Then run the script again.

Run

To start detecting images, run this in root folder. (windows user must run this via wsl)

python3 start.py -f [filename]

If you want to detect images in batch (highly recommended if you have multiple images that want to be detected), add -b option

python3 start.py -b --img_path="/home/img/"