Build a local dashboard for indexing and labeling
Wrap indexing, labeling, threshold tuning and export in one local browser page, and handle the threading traps that come with it.
Newest first. Filtering and search run in your browser over the whole list, so there is no round trip.
9 guides
Wrap indexing, labeling, threshold tuning and export in one local browser page, and handle the threading traps that come with it.
Fit a classifier on frozen embeddings in about a second, correct the cases it is least sure about, and know when to stop.
The literal error messages from building a local image search tool, why each happened, and what fixed it.
Why CLIP cannot tell people apart, what to use instead, and how clustering discovers everyone in your library with almost no input.
A seven-part series on building a private image search tool that learns which photos you actually want.
Index a folder of photos with CLIP and search it by typing what you remember seeing, with no cloud and no API key.
Install CLIP, download the weights once, cut the network, and turn a folder of photos into a searchable index.
How a sentence gets compared to a photo with one matrix multiply, how to read the scores, and why text search plateaus.
Text search ranks by distance to one fixed point that sits in the wrong place, so fit a boundary on your own labels instead.