A trend starts from "Chain of Thought Prompting Elicits Reasoning in Large Language Models".

Intro to Reinforcement Learning (强化学习纲要）

Assessing syntactic abilities of BERT

Evaluating recurrent neural networks on predicting subject-verb agreement dependencies

A 1D analogue of the MNIST dataset for measuring spatial biases and answering Science of Deep Learning questions.

An implement of the paper of EDA for Chinese corpus.中文语料的EDA数据增强工具。NLP数据增强。论文阅读笔记。

UDA(Unsupervised Data Augmentation) implemented by pytorch

Pretrained BERT model for analysing COVID-19 Twitter data

This is a simple Python package for calculating a variety of lexical diversity indices

COVID-19 Question Dataset from the paper "What Are People Asking About COVID-19? A Question Classification Dataset"

TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/

MNIST converted to PNG format

jiant is an nlp toolkit

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

⚡ A Fast, Extensible Progress Bar for Python and CLI

Functions for pre-processing large images

Paper List for Style Transfer in Text

Scaling and Benchmarking Self-Supervised Visual Representation Learning

Code for "Generative Image Translation for Data Augmentation in Colorectal Histopathology Images" full paper at ML4H Workshop at NeurIPS 2019.

A High-Throughput Workflow for Preprocessing, Deep Learning Analytics and Interpretation in Digital Pathology

Whole Slide Digital Pathology Image Tissue Localization

Digital Pathology Whole Slide Image Analysis Toolbox

Tensorflow implementation for learning an image-to-image translation without input-output pairs. https://arxiv.org/pdf/1703.10593.pdf

Deepfakes Software For All

Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.

Code for the paper 'Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic Image Colorization with Simultaneous Classification'.

Seq2seq chatbot with attention and anti-language model to suppress generic response, option for further improve by deep reinforcement learning.