-
Notifications
You must be signed in to change notification settings - Fork 465
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
[Question] Could we consider utilizing the index_mapping
feature of IntNBitTableBatchedEmbeddingBagsCodegen
to enhance the inference performance of QuantManagedCollisionEmbeddingCollection
#2669
Comments
hi, @kausv @dstaay-fb @dracifer @sarckk @TroyGarden @PaulZhang12 can you see this problem? |
Hi, @henrylhtsang @joshuadeng @YLGH can you see this problem? |
Hey, I've relayed this information to a member on our team who wrote this - they should get back to you shortly |
thx! |
@tiankongdeguiji I think that could work. |
thx!. The default value of the index_mapping is -1, whereas the default value of the _mch_remap is |
hi, @kausv , can you see the default value problem as said above? |
In the current version of _mch_remap within the
MCHManagedCollisionModule
, thesearchsorted
method is used for remapping, which may not be the most efficient approach for inference. To enhance the inference performance of theQuantManagedCollisionEmbeddingCollection
, could we consider using theindex_mapping
feature ofIntNBitTableBatchedEmbeddingBagsCodegen
?The text was updated successfully, but these errors were encountered: