-
-
Notifications
You must be signed in to change notification settings - Fork 332
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
Address error on /learn
after change of embedding model
#870
Conversation
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.
@srdas
Thanks for fixing this. It might be better to make the default in learn_dir
as False, which is the most used option. Also, it seems like that relearn
which indexes the files in case the embedding model is changed, should save the all_files
option initially passed from the user.
For an immediate fix, let's make all_files default as False, long term let's figure out how to persist the all_files option passed by the user, so that relearn works in the intended way when model is changed.
@3coins As suggested, the default in |
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.
Looks good, should be ok to merge after the rebase.
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
0357d96
to
7a1f4f0
Compare
…#870) * Address error on `/learn` after change of embedding model. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Set `all_files` option to default to False * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Fixes Issue #869
The
all_files
parameter was not being passed to learn the RAG folder when the embedding provider is changed. This has now been fixed and tested. To test, an initial embedding model is used and/learn
is applied.(See Issue #869 for the error that is addressed below.)
Next, the embedding model is changed on the settings page and
/learn
is tested again to make sure it is working as expected, as shown: