-
Notifications
You must be signed in to change notification settings - Fork 67
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
Migrate feature diff for NN Descent from RAFT to cuVS #421
Migrate feature diff for NN Descent from RAFT to cuVS #421
Conversation
void build(raft::resources const& res, | ||
index_params const& params, | ||
raft::device_matrix_view<const float, int64_t, raft::row_major> dataset, | ||
index<uint32_t>& index); |
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 doesn't nn-descent return the built index like all the other index types?
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 does, there's an API for that as well. We need this API as well especially for CAGRA because it needs to own the knn graph that it sends to NN Descent. For that, it needs to construct an index first and that's why we need this API.
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.
But the usage docs for this function don't reflect that. The whole idea behind these build
functon is to abstract away the index so that the user (and anyone using the public APIs) don't need to think about the underlying index object. Instead of having the user construct the index object on their own, we shuold have them pass an optional knn graph into the build
function that it then uses when it constructs the underlying index instance underneath.
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.
Done in latest commit
index<uint32_t>& index); | ||
|
||
/** | ||
* @brief Build nn-descent Index with dataset in host memory |
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.
Make sure all of these are exposed through the docs.
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.
Yes, they are part of the doxygen group that is already in the docs source.
@@ -146,6 +189,125 @@ class AnnNNDescentTest : public ::testing::TestWithParam<AnnNNDescentInputs> { | |||
rmm::device_uvector<DataT> database; | |||
}; | |||
|
|||
template <typename DistanceT, typename DataT, typename IdxT> | |||
class AnnNNDescentBatchTest : public ::testing::TestWithParam<AnnNNDescentBatchInputs> { |
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.
Do we have this consistently passing yet?
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.
No, that's a follow-on item. The tests are disabled for now both here and in RAFT.
/merge |
This PR is an amalgamation of the diff of 3 PRs in RAFT:
This PR also addresses part 1 and 2 of #419, closes #391 and makes CAGRA use the compiled headers of NN Descent, which seemed to have been a pending TODO
cuvs/cpp/src/neighbors/detail/cagra/cagra_build.cuh
Lines 36 to 37 in 009bb8d
Also, batch tests are disabled in this PR due to issue rapidsai/raft#2450. PR #424 will attempt to re-enable them.