Name: João Lourenço Silva
Type: User
Company: Instituto Superior Técnico / INESC-ID
Bio: MSc in Electrical and Computer Engineering at Instituto Superior Técnico. Master thesis on deep learning and computer vision applied to medical image analysis.
Twitter: JoaoLSilva99
Location: Lisbon
João Lourenço Silva's Projects
Python implementation of algorithms from Russell And Norvig's "Artificial Intelligence - A Modern Approach"
Unofficial implementation of the ImageNet, CIFAR 10 and SVHN Augmentation Policies learned by AutoAugment using pillow
Google Brain AutoML
Decoder architecture based on the UNet++. Combining residual bottlenecks with depthwise convolutions and attention mechanisms, it outperforms the UNet++ in a coronary artery segmentation task, while being significantly more computationally efficient.
Neural network graphs and training metrics for PyTorch, Tensorflow, and Keras.
Machine learning metrics for distributed, scalable PyTorch applications.
AI Toolkit for Healthcare Imaging
A minimal Jekyll Theme to host your resume (CV)
Tensors and Dynamic neural networks in Python with strong GPU acceleration
The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.
Simple React App to make QCA annotations on x-ray coronary angiography images and corresponding and semantic segmentation masks.
Segmentation models with pretrained backbones. PyTorch.
Code and data for our paper "The Role of Syntactic Planning in Compositional Image Captioning", EACL 2021.
Datasets, Transforms and Models specific to Computer Vision
Simple web-based monitoring tool for your small GPU cluster.