Streaming data pipeline that learns from 47GB of historical eBird observations to generate synthetic bird sightings and detect anomalies using Java, Kafka Streams, and Docker.
-
Updated
Jul 18, 2025 - Python
Streaming data pipeline that learns from 47GB of historical eBird observations to generate synthetic bird sightings and detect anomalies using Java, Kafka Streams, and Docker.
Bird Identification System with TensorFlow & eBird API 0.2
A Home Assistant integration for bird sightings within an area you choose, with the notable ones, your year list and first sightings of the year, and the sightings on a map.
To associate your repository with the ebird-api topic, visit your repo's landing page and select "manage topics."