FFT methods: Rader and Bluestein. Implemented in Python.

A presentation/notebook that expresses my view on things making PyTorch efficient that targets researchers in AI and other domains.

Spectral Graph Pruning Against Over-squashing and Over-smoothing (NeurIPS 2024)

GNNs Getting ComFy: Community and Feature Similarity Guided Rewiring (ICLR 2025)

Harness for training/finding lottery tickets in PyTorch. With support for multiple pruning techniques and augmented by distributed training, FFCV and AMP.

Talking about stocks on the weekends while drinking some tea.

This is the public repository for my TFG (trabajo de fin de grado, end-of-degree project). It is based on DeepCube and DeepCubeA papers

A tensorflow convolutional neural network model to detect facial expressions.