matthijs aa1585cb21 * 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
2017-06-21 07:22:50 -07:00
..
2017-03-02 22:31:32 -08:00