set some options on how to copy to GPU More...
#include <GpuAutoTune.h>
Public Attributes | |
IndicesOptions | indicesOptions |
how should indices be stored on GpuIndexIVFPQ? | |
bool | useFloat16CoarseQuantizer |
is the coarse quantizer in float16? | |
bool | useFloat16 |
bool | usePrecomputed |
use precomputed tables? | |
long | reserveVecs |
reserve vectors in the invfiles? | |
int | verbose |
set some options on how to copy to GPU
Definition at line 31 of file GpuAutoTune.h.
bool faiss::gpu::GpuClonerOptions::useFloat16 |
for GpuIndexIVFFlat, is storage in float16? for GpuIndexIVFPQ, are intermediate calculations in float16?
Definition at line 38 of file GpuAutoTune.h.