Feng Qian's repositories
multi-label-bert-classification
Multi-label Bert Classification with focal loss weighting, auto cross-label data synthesis, adding exclude loss part among specific labels, upsampling, robust mean over all positive or negative loss, generating very fast inference-time model, etc.
technical-skill-summary
My technical skill summary
algorithm-components
This library was designed for high performance pipeline of feature transform and model prediction as a consistent solution for both online and offline scenes.
preprocess_tabular_data
Auto inspect tabular data and transform data with PMML
tensorflow_highlevelapi_example
Tensorflow high level API example
angel
A Flexible and Powerful Parameter Server for large-scale machine learning
audio_to_midi_melodia
Extract the melody from an audio file and export to MIDI
chinese_ocr
CTPN + DenseNet + CTC based end-to-end Chinese OCR implemented using tensorflow and keras
ChordNova
ChordNova is a powerful open-source chord progression analysis plus generation software with unprecedentedly detailed control over chord trait parameters, that is way above mainstream softwares. Runs on multiple OS (currently Windows and Linux). | 智弦（ChordNova）是清华大学沈智云和星海音乐学院陈文戈共同开发的一款免费开源、功能强大的和弦进行自动生成软件。该软件提供前所未有的特征参数细节控制，远超以三度叠置为基础的主流软件。
DB
A PyToch implementation of "Real-time Scene Text Detection with Differentiable Binarization".
deep-text-recognition-benchmark
Text recognition (optical character recognition) with deep learning methods.
DifferentiableBinarization
DB (Real-time Scene Text Detection with Differentiable Binarization) implementation in Keras and Tensorflow
fastai
The fast.ai deep learning library, lessons, and tutorials
ICDAR2019-ArT-Recognition-Alchemy
PKU Team Zero's code for participation in ICDAR2019 ArT Recognition track (Champion)
Image-Downloader
Download images from Google, Bing, Baidu. 谷歌、百度、必应图片下载.
ismir2017_chords
ISMIR 2017: structured training for large vocab chord recognition
keras-ocr
A packaged and flexible version of the CRAFT text detector and Keras CRNN recognition model.
libsnark
C++ library for zkSNARKs
melosynth
Synthesize a continuous pitch sequence
OTR
Optical table recognition - recognize tables in scan images using OpenCV
PaddleOCR
基于飞桨的OCR工具库，包含总模型仅8.6M的超轻量级中文OCR，单模型支持中英文数字组合识别、竖排文本识别、长文本识别。同时支持多种文本检测、文本识别的训练算法。
qf6101.github.io
blog source file
scaper
A library for soundscape synthesis and augmentation
SealUtil
印章生成工具：使用Java Graphics2D生成各类圆形/椭圆公章、私章图片
SynthText
Code for generating synthetic text images as described in "Synthetic Data for Text Localisation in Natural Images", Ankush Gupta, Andrea Vedaldi, Andrew Zisserman, CVPR 2016.
TextRecognitionDataGenerator
A synthetic data generator for TEXT DETECTION (as opposed to text recognition)
Transformer
Transformer seq2seq model, program that can build a language translator from parallel corpus
you-get
:arrow_double_down: Dumb downloader that scrapes the web
