]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
ggml : unit test for quantization functions (#953)
authorunbounded <redacted>
Sat, 22 Apr 2023 09:10:39 +0000 (11:10 +0200)
committerGitHub <redacted>
Sat, 22 Apr 2023 09:10:39 +0000 (12:10 +0300)
commit5f939498d517b4dddbe904f202e895a3ecfb9dc4
tree010b9312921f4234bda24d3ff10aba86e1d3b6ac
parent36b4f7e06406eed8a605cc9f2921d9244ef6a8e5
ggml : unit test for quantization functions (#953)

* Unit test for quantization functions

Use the ggml_internal_get_quantize_fn function to loop through all
quantization formats and run a sanity check on the result.

Also add a microbenchmark that times these functions directly without
running the rest of the GGML graph.

* test-quantize-fns: CI fixes

Fix issues uncovered in CI
 - need to use sizes divisible by 32*8 for loop unrolling
 - use intrinsic header that should work on Mac

* test-quantize: remove

Per PR comment, subsumed by test-quantize-fns

* test-quantize: fix for q8_0 intermediates
tests/CMakeLists.txt
tests/test-quantize-fns.cpp [new file with mode: 0644]
tests/test-quantize-perf.cpp [new file with mode: 0644]
tests/test-quantize.c [deleted file]