From 5b09797321430f08caf0473143a962916ab2ea89 Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Sat, 9 Mar 2024 15:53:59 +0200 Subject: ggml : remove old quantization functions (#5942) * ggml : remove old quantization functions ggml-ci * ggml : simplify ggml_quantize_chunk ggml-ci * ggml : restrict correctness ggml-ci * ggml : remove hist data from the quantization API ggml-ci * tests : remove hist usage in test-backend-ops ggml-ci * vulkan : remove hist and fix typo --- examples/benchmark/benchmark-matmult.cpp | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'examples/benchmark/benchmark-matmult.cpp') diff --git a/examples/benchmark/benchmark-matmult.cpp b/examples/benchmark/benchmark-matmult.cpp index e89f3de2..47cb16c6 100644 --- a/examples/benchmark/benchmark-matmult.cpp +++ b/examples/benchmark/benchmark-matmult.cpp @@ -189,12 +189,10 @@ int main(int argc, char ** argv) { int32_t nelements = sizex*sizey; - std::vector hist_cur(1 << 4, 0); - // Set up a the benchmark matrices // printf("Creating new tensor q11 & Running quantize\n"); struct ggml_tensor * q11 = ggml_new_tensor_2d(ctx, qtype, sizex, sizey); - ggml_quantize_chunk(qtype, (const float *) m11->data, q11->data, 0, nelements/m11->ne[0], m11->ne[0], hist_cur.data(), nullptr); + ggml_quantize_chunk(qtype, (const float *) m11->data, q11->data, 0, nelements/m11->ne[0], m11->ne[0], nullptr); // Set up a the compute graph // printf("Creating new tensor q31\n"); @@ -207,7 +205,7 @@ int main(int argc, char ** argv) { // Set up a second graph computation to make sure we override the CPU cache lines // printf("Creating new tensor q12 & Running quantize\n"); struct ggml_tensor * q12 = ggml_new_tensor_2d(ctx, qtype, sizex, sizey); - ggml_quantize_chunk(qtype, (const float *) m12->data, q12->data, 0, nelements/m12->ne[0], m12->ne[0], hist_cur.data(), nullptr); + ggml_quantize_chunk(qtype, (const float *) m12->data, q12->data, 0, nelements/m12->ne[0], m12->ne[0], nullptr); // printf("Creating new tensor q32\n"); struct ggml_tensor * q32 = ggml_mul_mat(ctx, q12, m2); -- cgit v1.2.3