Hi,
I have created Text Searcher.tflite model in python and use this in my android enviroment.
As the documentation states; "New items can be added simply re-building the index. "
This index file is stored within the metadata of the tflite model.
Is there any way of appending new embeddings to the metadata and rebuild this index on android?
When inspecting the metadata;
{
"associated_files": [
{
"name": "on_device_scann_index.ldb",
"description": "On-device Scann Index file with LevelDB format.",
"type": "SCANN_INDEX_FILE"
}
]
}
The ScaNN layer exposes certain index building functionality but how can I execute this given the model and new entries?
Thanks for your help!
QNTN