-
-
Couldn't load subscription status.
- Fork 47
Open
@aguaviva
Description
I index the images using:
(base) $ ~/repos/clip.cpp/build$ bin/image-search-build -m ./models/CLIP-ViT-B-32-laion2B-s34B-b79K_ggml-model-f16.gguf ../tests/
clip_model_load: description: two-tower CLIP model
clip_model_load: GGUF version: 2
clip_model_load: alignment: 32
clip_model_load: n_tensors: 397
clip_model_load: n_kv: 24
clip_model_load: ftype: f16
clip_model_load: text_encoder: 1
clip_model_load: vision_encoder: 1
clip_model_load: model size: 289.06 MB
clip_model_load: metadata size: 0.13 MB
clip_model_load: 24 MB of memory allocated
main: starting base dir scan of '../tests/'
main: processing 2 files in ''
.main: 2 images processed and indexed
And here are the distances as computed in the readme.md the problem is that I am getting different values
(base) $ ~/repos/clip.cpp/build$ bin/image-search apple
clip_model_load: description: two-tower CLIP model
clip_model_load: GGUF version: 2
clip_model_load: alignment: 32
clip_model_load: n_tensors: 397
clip_model_load: n_kv: 24
clip_model_load: ftype: f16
clip_model_load: text_encoder: 1
clip_model_load: vision_encoder: 1
clip_model_load: model size: 289.06 MB
clip_model_load: metadata size: 0.13 MB
clip_model_load: 24 MB of memory allocated
search results:
distance path
0.711100 ../tests//red_apple.jpg
0.779483 ../tests//white.jpg
How could I tell what is right?
Metadata
Metadata
Assignees
Labels
No labels