clip-as-service
clip-as-service copied to clipboard
updated readme
updated readme to show command for building docker image locally.
Ok, I see. You want to build CUDA
image. May I ask you to add the building CLI in our document instead of README?
Codecov Report
Merging #903 (ead4e60) into main (f4eed3b) will decrease coverage by
20.59%
. The diff coverage isn/a
.
@@ Coverage Diff @@
## main #903 +/- ##
===========================================
- Coverage 82.73% 62.14% -20.59%
===========================================
Files 23 23
Lines 1564 1564
===========================================
- Hits 1294 972 -322
- Misses 270 592 +322
Flag | Coverage Ξ | |
---|---|---|
cas | 62.14% <ΓΈ> (-20.59%) |
:arrow_down: |
Flags with carried forward coverage won't be shown. Click here to find out more.
see 12 files with indirect coverage changes
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.
Hi @numb3r3
Yeah that's fine. I wanted to use it as an executor in a jina flow
. The documentation wasn't clear that was the correct way to locally build for the flow
use case.
@themantalope For your purpose, you can directly use the containerized executor from the hub jinahub+docker://CLIPTorchEncoder
(https://cloud.jina.ai/executor/gzpbl8jh):
jtype: Flow
executors:
- uses: jinahub+docker://CLIPTorchEncoder
In the above example, the executor will be a docker container in a flow.
But you are right, the document is not clear and misleading. Thank you for pointing that out.
@numb3r3 the other reason why I wanted to build is because I needed to fork the code and add debug statements. Even on JINA_LOG_LEVEL=DEBUG
I was getting errors from the indexer
stating that a Doc.embedding
was None
. I was trying to figure out which piece of data it was etc.