-
matthijs authored
* moved most FAISS_ASSERT calls to C++ exceptions, and adjusted memory allocation to avoid mem leaks * added an IndexIVFScalarQuantizer type that offers an intermediate compression between IVFFlat and IVFPQ * support removal of indices in IndexIDMap / IndexFlat combination * various fixes in GPU code
784e2fac
| Name |
Last commit
|
Last update |
|---|---|---|
| .. | ||
| fp16_emu.cu | ||
| fp16_emu.cuh |