-
Notifications
You must be signed in to change notification settings - Fork 72
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add vertex feature store integration #250
Conversation
...ertexai/tests/integration_tests/vectorstores/feature_store/test_feature_store_bf_executor.py
Outdated
Show resolved
Hide resolved
libs/vertexai/langchain_google_vertexai/vectorstores/feature_store/executors.py
Outdated
Show resolved
Hide resolved
/gcbrun |
1 similar comment
/gcbrun |
127d024
to
b96cecf
Compare
/gcbrun |
/gcbrun |
1 similar comment
/gcbrun |
In integration tests could you please add a check for already created datasets in BQ. |
@gmogr Added teardown for feature view. |
/gcbrun |
/gcbrun |
gca_resource = self._online_store.gca_resource | ||
endpoint = gca_resource.dedicated_serving_endpoint.public_endpoint_domain_name | ||
return FeatureOnlineStoreServiceClient( | ||
client_options={"api_endpoint": endpoint} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it looks like telemetry is missing here too
self._bq_client = bigquery.Client( | ||
project=self.project_id, | ||
location=self.location, | ||
credentials=self.credentials, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
same here, client_info is not set
authentication. | ||
|
||
To use, you need the following packages installed: | ||
google-cloud-bigquery |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can we hint a specific langchain-google-vertexai
dep group instead, please?
|
||
def sync(self): | ||
"""Sync the data from the Big Query source into the Executor source""" | ||
self._df = self._query_table_to_df() |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can we at least add a warning if the amount of records it too large for an in-memory bfs, please?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
+1
self._env_config = env_config | ||
|
||
|
||
class BruteForceExecutor(BaseExecutor): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why do we need our own in-memory implementation if LC already has a few? can we just write an adapter that will load data into it?
The FeatureStore aims to facilitate similarity search using different | ||
methodologies on Google Cloud including Big Query, Feature Store and a | ||
local bruteforce search engine. | ||
Big Query is the data source of truth and also the default search |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: BigQuery
one word instead of Big Query
Authors: @lspataroG @eliasecchig
Add VertexAI Langchain integration
Example usage