conditional-neural-process

Code for the Neural Processes website and replication of 4 papers on NPs. Pytorch implementation.

Implementation of the Convolutional Conditional Neural Process

Source codes for "Improved Few-Shot Visual Classification" (CVPR 2020), "Enhancing Few-Shot Image Classification with Unlabelled Examples" (WACV 2022), and "Beyond Simple Meta-Learning: Multi-Purpose Models for Multi-Domain, Active and Continual Few-Shot Learning" (SSRN Electronic Journal)

Source codes for "Improved Few-Shot Visual Classification" (CVPR 2020), "Enhancing Few-Shot Image Classification with Unlabelled Examples" (WACV 2022), and "Beyond Simple Meta-Learning: Multi-Purpose Models for Multi-Domain, Active and Continual Few-Shot Learning" (Neural Networks 2022 - in submission)

PyTorch models for AV1 artifact removal. Features a universal CRF-Preset conditional model (FiLM-based U-Net) and ultra-light Nano models (U-Net, ResNet, ...) for fast real-time video restoration (different model for each CRF bucket).

Official code for "Adaptive Conditional Quantile Neural Processes" (UAI 2023)

Implementation of a generic wrapping algorithm that can enable any meta-learning algorithm to perform well on unsegmented time-series data.

Leak-free neural surrogates of the marginal stability of an oscillatory Taylor-Couette flow in a UCM fluid: conditional neural process, distilled transformer, SARL and MARL refiners, with the Floquet database and the evaluation. Extends Hayani Choujaa et al., Phys. Fluids 33, 074105 (2021).