google-deepmind/neural-processes
Google DeepMind's notebook implementations of Neural Process variants (CNPs, NPs, ANPs) for meta-learning and function approximation.
Not currently ranked — collecting fresh signals.
star history
This repository contains Jupyter notebook implementations of three neural process model variants: Conditional Neural Processes (CNPs), Neural Processes (NPs), and Attentive Neural Processes (ANPs). These are deep learning architectures that combine neural networks with stochastic processes for meta-learning tasks, enabling function approximation from limited context data. The code runs on TensorFlow and can be executed in the browser via Google Colab or locally with Jupyter.
Frequently asked
- What is google-deepmind/neural-processes?
- Google DeepMind's notebook implementations of Neural Process variants (CNPs, NPs, ANPs) for meta-learning and function approximation.
- Is neural-processes open source?
- Yes — google-deepmind/neural-processes is open source, released under the Apache-2.0 license.
- What language is neural-processes written in?
- google-deepmind/neural-processes is primarily written in Jupyter Notebook.
- How popular is neural-processes?
- google-deepmind/neural-processes has 1k stars on GitHub.
- Where can I find neural-processes?
- google-deepmind/neural-processes is on GitHub at https://github.com/google-deepmind/neural-processes.