-
Notifications
You must be signed in to change notification settings - Fork 2.8k
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
[Feature]: Support Double, Float16 and BF16 vectors #22837
Comments
/assign |
@xiaofan-luan could I ask if it is wrong to convert the embedding to a float32, which I think has better numerical performance on most CPU unless hardware support exists? Or, is the purpose of this issue to support storage of such formats assuming that the compute nodes have the correct compute (e.g. GPU or the right Xeon chipset) to handle operations in those datatypes? If so, do we need to implement fallback by e.g. emulation or casting when the appropriate compute support is missing? Pytorch handles by autocasting. |
Duplicate: #22132 |
btw, bfloat16 does not exist on faiss: https://github.com/facebookresearch/faiss/wiki/How-to-make-Faiss-run-faster, and I believe not in Annoy or HNSWLib either But it supports float16 and we can compile it back in: #2828 |
Welcome @jon-chuang , You can implement float16 first, we can discuss about bf16 later. 😄 |
/unassign @jon-chuang |
We can break down the steps into the following:
|
…e as BinaryVector (milvus-io#33760) Issue: milvus-io#22837 Signed-off-by: Cai Yudong <yudong.cai@zilliz.com>
…pe (milvus-io#33625) issue: milvus-io#22837 pr: milvus-io#33624 Signed-off-by: xianliang.li <xianliang.li@zilliz.com>
issue: #22837 - fix byte size wrong for binary vectors - fix the expect/actual error msg Signed-off-by: chasingegg <chao.gao@zilliz.com>
…-io#33377) related milvus-io#22837 Signed-off-by: chasingegg <chao.gao@zilliz.com>
issue:milvus-io#22837 related milvus-io#33575 Signed-off-by: cqy123456 <qianya.cheng@zilliz.com>
…e as BinaryVector (milvus-io#33760) Issue: milvus-io#22837 Signed-off-by: Cai Yudong <yudong.cai@zilliz.com>
issue:milvus-io#22837 related:milvus-io#33653 Signed-off-by: cqy123456 <qianya.cheng@zilliz.com>
issue: milvus-io#22837 contain milvus-io#33625 milvus-io#33867 milvus-io#33911 which already merged to 2.4 branch Signed-off-by: chasingegg <chao.gao@zilliz.com> Co-authored-by: foxspy <xianliang.li@zilliz.com>
issue: milvus-io#22837 - fix byte size wrong for binary vectors - fix the expect/actual error msg Signed-off-by: chasingegg <chao.gao@zilliz.com>
issue: milvus-io#22837 related: milvus-io#33878 Signed-off-by: cqy123456 <qianya.cheng@zilliz.com>
…move some cgo call (milvus-io#34102) issue: milvus-io#22837 related pr: milvus-io#34104 Signed-off-by: cqy123456 <qianya.cheng@zilliz.com>
Is there an existing issue for this?
Is your feature request related to a problem? Please describe.
There are many different vector types based on models.
So far what we received most is double, float16, BF16, double and BF16 is on top priority.
Anyone interested on it please help
Describe the solution you'd like.
No response
Describe an alternate solution.
No response
Anything else? (Additional Context)
No response
The text was updated successfully, but these errors were encountered: