metadata

license: mit
task_categories:
  - text-generation
  - image-classification
language:
  - km
  - en
tags:
  - ocr
  - khmerocr
  - font-printed-text-dataset
pretty_name: Khmer-Printed-Text-Dataset
size_categories:
  - 10K<n<100K

62k Khmer-English Printed Dataset

This repository contains a dataset of Khmer and English printed text images for training, validation, and testing. The dataset is stored in parquet format and managed using Git Large File Storage (LFS).

Installation

Prerequisites

Before cloning this repository, make sure you have Git LFS installed:

Install Git LFS

Linux/macOS:

curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
sudo apt install git-lfs

Windows: Download and install Git LFS from git-lfs.github.com

Clone the Repository

git clone https://github.com/SoyVitou/62k-images-khmer-printed-dataset.git
cd 62k-images-khmer-printed-dataset

Pull LFS Files

git lfs install
git lfs pull

Dataset Files

trainset.parquet (1.92 GB) - Training set with Khmer and English printed text.
validset.parquet - Validation set.
testset.parquet (238 MB) - Testing set with Khmer and English printed text.

Usage

You can load the dataset using Python with Pandas:

import pandas as pd

# Load training set
df_train = pd.read_parquet("trainset.parquet")
print(df_train.head())

# Load validation set
df_valid = pd.read_parquet("validset.parquet")
print(df_valid.head())

# Load test set
df_test = pd.read_parquet("testset.parquet")
print(df_test.head())

License

[Specify your license here, e.g., MIT, Apache 2.0, etc.]

Contact

For any issues or inquiries, please contact [your email or GitHub profile].