Mel Augusto's starred repositories
cs-video-courses
List of Computer Science courses with video lectures.
Prompt-Engineering-Guide
🐙 Guides, papers, lecture, notebooks and resources for prompt engineering
tuning_playbook
A playbook for systematically maximizing the performance of deep learning models.
chatgpt-retrieval-plugin
The ChatGPT Retrieval Plugin lets you easily find personal or work documents by asking questions in natural language.
skill-icons
Showcase your skills on your Github readme or resumé with ease ✨
text-to-text-transfer-transformer
Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"
keep-a-changelog
If you build software, keep a changelog.
alignment-handbook
Robust recipes to align language models with human and AI preferences
LLM-eval-survey
The official GitHub page for the survey paper "A Survey on Evaluation of Large Language Models".
Software-Engineering
Learning Software Engineering :turtle:
mengzi-zero-shot
NLU & NLG (zero-shot) depend on mengzi-t5-base-mt pretrained model
conceitos-basicos-NLP
Aulas de conceitos básicos de Processamento de Linguagem Natural oferecida no Discord aberto no Turing USP
torchserver-docker-compose
Run Torch Serve with Docker using a docker-compose.yml
covid19
Collection of 3,925,366 tweets and 18,413 online news around the online discussion about COVID-19 in Brazil. The data from Twitter were collected through Twitterscraper Python library and we considered a set of keywords in Portuguese regarding to COVID-19. In order to facilitate the identification of tweets that have hashtags, media and retweets for researchers or data enthusiasts, we created three specific datasets for each of these categories. The news on COVID-19 was collected from the UOL portal, the most popular Brazilian website. All the data were gathered from January to May, 2020. These dataset can attract the attention from communities such as data science, social science, natural language processing, tourism, infodemiology, and public health.
FakeOnlineTCC-Corpus
FakeOnline is a corpus composed of true and fake news in Brazilian Portuguese.