nearle_daily/catalogue_backend
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

image vector dimensionality reduction

Browse Source
This commit is contained in:
sriram
2026-09-17 14:21:48 +05:30
parent deae694a1f
commit afa0bfa743
11 changed files with 907 additions and 183 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
.env.example
View File

@@ -314,12 +314,17 @@ USE_PLAYWRIGHT_FALLBACK=true
MIN_IMAGE_BYTES=3000
# img_vector: a 32x32 RGB pixel thumbnail of each product's primary image,
# stored on its brand table and computed on a background thread after every
# catalog write (app/services/image_vector.py). Existing rows are filled by
# img_vector: a MobileNetV3-Small embedding (1024 floats, L2-normalised) of
# each product's primary image, stored on its brand table and computed on a
# background thread after every catalog write (app/services/image_vector.py,
# model in app/services/image_embedder.py). Existing rows are filled by
# `python -m scripts.backfill_image_vectors --all --apply`. Set false to stop
# the API process from downloading images at all.
ENABLE_IMAGE_VECTORS=true
# Where the .tflite lives. The default is inside app/ (shipped with the image
# and never volume-mounted); only override to point at a different file.
#IMAGE_EMBED_MODEL_PATH=/app/app/services/models/mobilenet/mobilenet_v3_small_embedder.tflite
#IMAGE_EMBED_NUM_THREADS=2
# Product SKU: try a live web search for a real marketplace product ID
# (Amazon ASIN, Flipkart PID, etc.) before falling back to an internal SKU.