Jinpeng

deep_ocr. make a better chinese character recognition OCR than tesseract

kdd2017. kdd2017 travel time competition rank 28/3574

gpu_shared_platform. An opensource platform to share your GPU at home

tesseract-trainer-docker. train tesseract with font files

web_template. auto define crawler

deep_ocr_cpu_docker. Make a cpu docker version for deep_ocr

deep_ocr_fonts. fonts used for deep_ocr

tesseract-trainer. tesseract-trainer

chinese_text_dataset. for training lstm

sync_phantomjs. PhantomJS is a HEADLESS browser, but it exists a problem about synchronization problem. This project builds a framework to solve the synchronization problem.

docker-image-reverse-ssh-tunnel. reverse ssh tunnel container

deep_ocr_trained_models. already trained models for the project deep_ocr

pystatsml. Statistics and Machine Learning in Python

extract_words. This module is used to extract words. e.g. price, sku, words, from pages.

glproto. Because of the network problem, I clone one into my repo.