Light

liuzhuang1024 / sam Goto Github PK

View Code? Open in Web Editor NEW

13.0 1.0 2.0 748 KB

Semantic Graph Representation Learning for Handwritten Mathematical Expression Recognition (ICDAR 2023)

License: GNU General Public License v3.0

Python 100.00%

handwrite-recognnize icdar2023 image-recognition recognition-algorithms formula-recognition

sam's Introduction

SAM

Semantic Graph Representation Learning for Handwritten Mathematical Expression Recognition (ICDAR 2023)

This is the official pytorch implementation of SAM[ICDAR'2023].

Zhuang Liu, Ye Yuan, Zhilong ji, Jingfeng Bai and Xiang Bai

Pipeline

Usage

You need config the config file and run train or inference after.

python train.py --config config_v2.yaml

Citation

@misc{liu2023semantic,
    title={Semantic Graph Representation Learning for Handwritten Mathematical Expression Recognition},
    author={Zhuang Liu and Ye Yuan and Zhilong Ji and Jingfeng Bai and Xiang Bai},
    year={2023},
    eprint={2308.10493},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Recommendation

Some other excellent open-sourced HMER algorithms can be found here:

WAP[PR'2017] DWAP-TD[ICML'2020] BTTR[ICDAR'2021] ABM[AAAI'2022] SAN[CVPR'2022] CoMER[ECCV'2022] CAN[ECCV'2022]

sam's People

Contributors

Stargazers

Watchers

Forkers

jeffer0323 zzy6660

sam's Issues

您好，能够提供训练模型，"checkpoints/can-l2-context-word_2022-12-08-10-19_decoder-Decoder_v3/2014_can-l2-context-word_2022-12-08-10-19_decoder-Decoder_v3_WordRate-0.9037_ExpRate-0.5720_179.pth"

您好，能否提供一下代码中的权重文件（checkpoints/can-l2-context-word_2022-12-08-10-19_decoder-Decoder_v3/2014_can-l2-context-word_2022-12-08-10-19_decoder-Decoder_v3_WordRate-0.9037_ExpRate-0.5720_179.pth）

您好，按照代码中的流程训练后，训练50轮word score为0.162 ExpRate为0.00，这是为什么呢

ZeroDivisionError: division by zero

when i run. py, some errors occurred in training.py：

File "train.py", line 106, in
eval_loss, eval_word_score, eval_expRate = eval(params, model, epoch, loader)
File "/home/SAM/training.py", line 192, in eval
return loss_meter.mean, word_right / length, exp_right / cal_num
ZeroDivisionError: division by zero

I found these four parameters: word_right, length, exp_right, cal_num are all 0

你好，能够提供一下权重模型

Why WRate is high but ERate is low?

I have trained 25 epoch, the WRate reached 0.778 but ERate is only 0.010 ?
Below is my configuration:
lr: 0.9
type: l2
batch_size: 8
workers: 4
use_flag: True
context_loss: True
word_state_loss: True

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.