Hugo Sousa's repositories
py_heideltime
Python wrapper for HeidelTime temporal tagger.
elmundo_scraper
Scrape news from ElMundo.
generate_fruit
Test diferent generative models on the Fruits 360 dataset.
gpt-fast
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
professor_heideltime
Create a multilingual corpus weakly labeled with HeidelTime.
Italian-Crime-News
A dataset from the Gazzetta di Modena newspaper about crime events in the province of Modena from 2011 to 2021 https://gazzettadimodena.gelocal.it/.
kannada_mnist
A deep CNN to classify the Kannada MNIST dataset.
langchain
⚡ Building applications with LLMs through composability ⚡
llama
Inference code for LLaMA models
lm-evaluation-harness-pt
A framework for few-shot evaluation of language models for Portuguese.
natural_questions_parser
Parser for the Natural Questions corpus.
publico
An European Portuguese news dataset translated to several languages.
publico_scraper
Scrape news from Jornal Público.
torchtune
A Native-PyTorch Library for LLM Fine-tuning
tutorials
PyTorch tutorials.