add skip_storage flag to HNSW #3487
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
This pull request was exported from Phabricator. Differential Revision: D57911060 |
alexanderguzhva
suggested changes
May 29, 2024
faiss/impl/index_write.cpp
Outdated
| WRITE1(n4); | ||
| } else { | ||
| write_index(idxhnsw->storage, f); | ||
| } | ||
| write_index(idxhnsw->storage, f); |
There was a problem hiding this comment.
writing the storage twice, dupilcate rows 770 and 772
junjieqi
reviewed
May 29, 2024
| @@ -251,7 +249,8 @@ void hnsw_search( | |||
| const SearchParameters* params_in) { | |||
| FAISS_THROW_IF_NOT_MSG( | |||
| index->storage, | |||
| "Please use IndexHNSWFlat (or variants) instead of IndexHNSW directly"); | |||
| "No storage index, please use IndexHNSWFlat (or variants) " | |||
| "instead of IndexHNSW directly"); | |||
There was a problem hiding this comment.
When do we prefer to using IndexHNSW and when should we use IndexHNSWFlat?
| @@ -531,7 +530,10 @@ Index* read_index(IOReader* f, int io_flags) { | |||
| Index* idx = nullptr; | |||
| uint32_t h; | |||
| READ1(h); | |||
| if (h == fourcc("IxFI") || h == fourcc("IxF2") || h == fourcc("IxFl")) { | |||
| if (h == fourcc("null")) { | |||
There was a problem hiding this comment.
When do we see the caller provide null ?
| @@ -765,6 +763,12 @@ void write_index(const Index* idx, IOWriter* f) { | |||
| WRITE1(h); | |||
| write_index_header(idxhnsw, f); | |||
| write_HNSW(&idxhnsw->hnsw, f); | |||
| if (io_flags & IO_FLAG_SKIP_STORAGE) { | |||
| uint32_t n4 = fourcc("null"); | |||
| WRITE1(n4); | |||
There was a problem hiding this comment.
is it gonna to output in stdout?
| np.testing.assert_array_equal(Dnew, Dref) | ||
| np.testing.assert_array_equal(Inew, Iref) | ||
|
|
||
| if False: |
There was a problem hiding this comment.
where do we define the False?
|
format & internal linter |
Summary: Sometimes it is not useful to serialize the storage index along with a HNSW index. This diff adds a flag that supports skipping the storage of the index. Searchign and adding to the index is not possible until a storage index is added back in. Reviewed By: junjieqi Differential Revision: D57911060
|
This pull request was exported from Phabricator. Differential Revision: D57911060 |
|
This pull request has been merged in bf73e38. |
tarang-jain
pushed a commit
to tarang-jain/faiss
that referenced
this pull request
Jul 5, 2024
Summary: Pull Request resolved: facebookresearch#3487 Sometimes it is not useful to serialize the storage index along with a HNSW index. This diff adds a flag that supports skipping the storage of the index. Searchign and adding to the index is not possible until a storage index is added back in. Reviewed By: junjieqi Differential Revision: D57911060 fbshipit-source-id: 5a4ceee4a8f53f6f746df59af3942b813a99c14f
abhinavdangeti
pushed a commit
to blevesearch/faiss
that referenced
this pull request
Jul 12, 2024
Summary: Pull Request resolved: facebookresearch#3487 Sometimes it is not useful to serialize the storage index along with a HNSW index. This diff adds a flag that supports skipping the storage of the index. Searchign and adding to the index is not possible until a storage index is added back in. Reviewed By: junjieqi Differential Revision: D57911060 fbshipit-source-id: 5a4ceee4a8f53f6f746df59af3942b813a99c14f
aalekhpatel07
pushed a commit
to aalekhpatel07/faiss
that referenced
this pull request
Oct 17, 2024
Summary: Pull Request resolved: facebookresearch#3487 Sometimes it is not useful to serialize the storage index along with a HNSW index. This diff adds a flag that supports skipping the storage of the index. Searchign and adding to the index is not possible until a storage index is added back in. Reviewed By: junjieqi Differential Revision: D57911060 fbshipit-source-id: 5a4ceee4a8f53f6f746df59af3942b813a99c14f
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Sometimes it is not useful to serialize the storage index along with a HNSW index. This diff adds a flag that supports skipping the storage of the index.
Searchign and adding to the index is not possible until a storage index is added back in.
Differential Revision: D57911060