Implement a custom tf.js loader

hotg-ai / rune

Rune provides containers to encapsulate and deploy edgeML pipelines and applications

Apache License 2.0

134 stars 15 forks source link

Implement a custom tf.js loader #386

Closed Ge-te closed 2 years ago

Ge-te commented 2 years ago

Added support for a zipped tf.js model. When encapsulating a zipped folder containing the models json and bin files in the rune, this IOHandler (TensorFlowZipRequest) loads it into memory without using local storage/indexedDb

Fixes #385.