Ziniu Yu

I tried to add warning messages to the client at the result gathering steps at first but changed to (image) preprocessing steps since in this way the process can proceed...

> I would like to evaluate the perf loss of this. Why dont we have the client say which doc Ids of a batch failed to be encoded? @JoanFM I...

Hello @sjcotto , glad to see you here! You can utilize the capability of [DocArray](https://docarray.jina.ai/) to store such information. For example, you can store product name, category to different tags...

Hi @sjcotto ! There are different ways to generate vector embeddings: 1. You can manually assign vector embedding [here](https://docarray.jina.ai/fundamentals/document/embedding/). 2. You can use Clip-as-Service to generate vector embedding using Clip...

Hi @sjcotto , could you provide more details like the error messages or the unexpected behaviors?

Hi @RaiAmanRai , you may take a look [here](https://clip-as-service.jina.ai/user-guides/server/#clip-model-config) and [here](https://docs.jina.ai/fundamentals/flow/yaml-spec/) For example, ```{yaml} jtype: Flow version: '1' with: port: 51000 executors: - name: clip_t uses: jtype: CLIPEncoder metas: py_modules:...

And also do some basic cleaning like `filename.strip()`

That sounds reasonable👍 Maybe we can make it an option to use the corresponding protocol if there is a specified one otherwise use the default one. One possible design is...

Hi @hsalehgh , please try again using Rosetta https://github.com/jina-ai/annlite/issues/116

server/clip_server/model/model.py This file is synced with https://github.com/mlfoundations/open_clip/blob/main/src/open_clip/model.py