-
Notifications
You must be signed in to change notification settings - Fork 371
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
Integration of dense embedding model Snowflake arctic-embed-m-v1.5 #2001
base: master
Are you sure you want to change the base?
Conversation
This is linked to resolving issue #1959 |
@@ -362,6 +364,43 @@ def encode(self, query: str, **kwargs): | |||
return self.get_embedding(inputs) | |||
else: | |||
return super().encode(query) | |||
|
|||
|
|||
class ArcticQueryEncoder(QueryEncoder): |
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 are there two different ArcticQueryEncoder
classes? Here and in encode._artic
?
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.
Yeah that is how the current version of pyserini does for all query encoders. Shivani and I were talking about this too. Maybe we could make a separate issue on this matter?
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.
@MXueguang ping on this?
@@ -413,6 +452,14 @@ def encode(self, query: str): | |||
return embeddings.flatten() | |||
else: | |||
return super().encode(query) | |||
|
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.
remove blank lines?
This update integrates the Snowflake arctic-embed-m-v1.5 dense embedding model into the Pyserini toolkit. A new class was developed to support the entire snowflake-series of models, offering a scalable foundation for future expansions, such as additional Snowflake embedding model integrations. This modification is intended to lay the groundwork for smoother integration of future models from the Snowflake series