Very recently, a variety of vision transformer architectures for dense prediction tasks have been proposed and they show that the design of spatial attention is critical to their success in these tasks
We all enjoy being part of tech (and non-tech) communities to learn, teach and share the things we know and want to know 🤓 . Here is a curated list of LATAM tech communities, organized by country > platform (Slack/Discord/Meetup/
This project provides the implementation for the CVPR 2021 paper "Scale-aware Automatic Augmentation for Object Detection". Scale-aware AutoAug provides a new search space and search metric to find effective data agumentation poli
This bot downloads and sends the audio when someone send a YouTube/SoundCloud/MixCloud link in the specified chats, there is a command /ping which makes the bot reply with a "ping" for checking if the bot is running.
CDIoU and CDIoU loss is like a convenient plug-in that can be used in multiple models. CDIoU and CDIoU loss have different excellent performances in several models such as Faster R-CNN, YOLOv4, RetinaNet and . There is a maximum A
It was an Unbelievable WeekEnd For me as I won Challenges#3 in UmojaHack Africa 2021 with my amazing Teammate @ASSAZZIN. this repository contains Code Solutions on how to get Top1 Place in Financial Resilience Challenge . Enjoy !
SeqIO is a library for processing sequential data to be fed into downstream sequence models. It uses tf.data.Dataset to create scalable data pipelines but requires minimal use of TensorFlow. In particular, with one line of code, t
Mixed precision training  is a technique that mixes the use of full and half precision floating point numbers during training to reduce the memory bandwidth requirements and improve the computational efficiency of a given model
Minimalistic open-source library for metric learning written in TensorFlow2, TF-Addons, Numpy, OpenCV(CV2) and Annoy. This repository contains a TensorFlow2+/tf.keras implementation some of the loss functions and miners. This repo
Pupil is a project in active, community driven development. Pupil Core mobile eye tracking hardware is accessible, hackable, and affordable. The software is open source and written in Python and C++ when speed is an issue.
Official PyTorch implementation for Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers, a novel method to visualize any Transformer-based network. Including examples for DETR, VQA.
TorchShard is a lightweight engine for slicing a PyTorch tensor into parallel shards. It can reduce GPU memory and scale up the training when the model has massive linear layers (e.g., ViT, BERT and GPT) or huge classes (millions