Adapt to v2.0.0 core #19
Conversation
alanthssss
force-pushed
the
adapt-to-v2.0.0
branch
from
89b41e3
to
182f45d
Compare
alanthssss
force-pushed
the
adapt-to-v2.0.0
branch
from
182f45d
to
15a7caa
Compare
There was a problem hiding this comment.
Thank you for porting this to 2.0 :)
Main comment is a question about sharding etc, would this be removed by the PR as it is now?
Maybe another question is if it makes sense to generalize the Executors implemented here and provide them as standalone Executors?
| with f: | ||
| f.index(inputs=DocumentArray([d1, d2, d3])) | ||
|
|
||
| with Flow.load_config( |
There was a problem hiding this comment.
Do we need this file in the PR? Looks like some local test to me which will not be used in the actual stress test? If you think its handy to have as some local test entry point, maybe generalize this and otherwise remove?
There was a problem hiding this comment.
@jacobowitz this will be removed after flow runs successfully, only for test purpose, shouldn't commit to the PR, good catch!
There was a problem hiding this comment.
thx for your review.
This PR is still in local progress.
| pods: | ||
| - name: segmenter | ||
| polling: any | ||
| shards: {{ JINA_SEGMENTER_SHARDS }} |
There was a problem hiding this comment.
why are we removing the sharding? Testing this as part of the stress test is one of the added values here I think?
There was a problem hiding this comment.
Maybe I should convert this PR to a draft one.
Let me fix the status of this PR.
|
@niuzs-alan things needs to be done (from the flow/executor level):
|
@bwanglzu