This project is presenting a new database with handwirtten Mongolian words.The training set of MHW database consists of 5000 words, each word is written 20 times, so it contains 100000 pieces of Mongolian words written by 200 different writers. The testing set I of MHW database consists of 1000 words, each word is written 5 times by the same writers with training set. The testing set II of MHW database consists of 939 words; each word is written 15 times written by 50 different writers with training set. The words in training and testing set have a few intersections. It is designed for training and testing recognition systems for handwritten Mongolian words. The MHW-database is available for the purpose of research.

usage instructions given at
https://sourceforge.net/p/mongolian-hw/wiki/Home/

Features

  • Mongolian
  • Handwritting recognition
  • dataset

Project Samples

MHW Screenshot 1

Project Activity

See All Activity >

Follow MHW

MHW Web Site

Other Useful Business Software
Host LLMs in Production With On-Demand GPUs Icon
Host LLMs in Production With On-Demand GPUs

NVIDIA L4 GPUs. 5-second cold starts. Scale to zero when idle.

Deploy your model, get an endpoint, pay only for compute time. No GPU provisioning or infrastructure management required.
Start Free

Additional Project Details

Registered

2017-03-01
Report inappropriate content