Jongsoo Park 7a837019a4 [caffe2] optimize 2/4-bit row-wise quantization (#387) ... Summary: Pull Request resolved: https://github.com/pytorch/FBGEMM/pull/387 Pull Request resolved: https://github.com/pytorch/pytorch/pull/39985 avx2 optimized 2/4-bit row-wise quantization/dequantization in perfkernels. This diff slightly change the numerics of quantization by multiplying with the inverse of scale instead of dividing with scale. Test Plan: In my devserver for i in 2 4 8; do echo $i; buck run mode/opt :fused_rowwise_nbit_conversion_bench -- --bit-rate=$i; done Before this diff 2-bit 3.35394 ms. 100%. FloatToFused2BitRowwiseQuantized 4-bit 3.60351 ms. 100%. FloatToFused4BitRowwiseQuantized 8-bit 0.434467 ms. 100%. FloatToFused8BitRowwiseQuantized After this diff 2-bit 0.606386 ms. 100%. FloatToFused2BitRowwiseQuantized 4-bit 0.446683 ms. 100%. FloatToFused4BitRowwiseQuantized 8-bit 0.4349 ms. 100%. FloatToFused8BitRowwiseQuantized Reviewed By: choudharydhruv, jianyuh Differential Revision: D22033195 fbshipit-source-id: d3a219e47b8345268d90a160c9314ed0d5b71467		2020-06-19 21:28:31 -07:00
..
fused_rowwise_nbit_conversion_bench.py	[caffe2] optimize 2/4-bit row-wise quantization (#387)	2020-06-19 21:28:31 -07:00
sparse_lengths_sum_nbit_benchmark.py	[caffe2] open source 2/4-bit SLS operators (#34903)	2020-03-17 22:55:10 -07:00