[ECAI 2023 Oral] PMAA: A Progressive Multi-scale Attention Autoencoder Model for High-Performance Cloud Removal from Multi-temporal Satellite Imagery

Let your GNOME desktop speak. Reads desktop notifications or selected text aloud with human voice using Piper. Uses a local LLM to summarize, translate, explain or proofread selected text.

Autoencoder.js

Latent Space uses a recurrent neural autoencoder that compresses audio streams, with embeddings specifically for music, to reduce audio latency.

Visualize high-dimensional data

Latent Space uses a recurrent neural autoencoder that compresses audio streams, with embeddings specifically for music, to reduce audio latency. Built on Zipcall.

Bring your code to the conversations you care about with the GitHub and Slack integration

A machine learning-based system for detecting anomalies in encrypted network traffic. Supports real-time analysis, multiple detection algorithms, and insightful visualizations.

PyTorch anomaly detection console for any tabular CSV. Originally scoped from an academic brief modeled on a maritime technology center's sonar-classification use case; independently rebuilt and generalized beyond the original dataset.