From d01bccde9f759b24449fdaa16306b406a50eb367 Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Tue, 18 Jul 2023 14:24:43 +0300 Subject: ci : integrate with ggml-org/ci (#2250) * ci : run ctest ggml-ci * ci : add open llama 3B-v2 tests ggml-ci * ci : disable wget progress output ggml-ci * ci : add open llama 3B-v2 tg tests for q4 and q5 quantizations ggml-ci * tests : try to fix tail free sampling test ggml-ci * ci : add K-quants ggml-ci * ci : add short perplexity tests ggml-ci * ci : add README.md * ppl : add --chunks argument to limit max number of chunks ggml-ci * ci : update README --- .gitignore | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to '.gitignore') diff --git a/.gitignore b/.gitignore index faec869..a23ac59 100644 --- a/.gitignore +++ b/.gitignore @@ -16,6 +16,8 @@ build/ build-em/ build-debug/ build-release/ +build-ci-debug/ +build-ci-release/ build-static/ build-cublas/ build-opencl/ @@ -25,9 +27,10 @@ build-no-accel/ build-sanitize-addr/ build-sanitize-thread/ out/ +tmp/ models/* -*.bin +models-mnt /main /quantize -- cgit v1.2.3 From 54e3bc76fed914f8d4a30a7a50c19867cccb1338 Mon Sep 17 00:00:00 2001 From: Jiří Podivín <66251151+jpodivin@users.noreply.github.com> Date: Fri, 21 Jul 2023 12:09:16 +0200 Subject: make : add new target for test binaries (#2244) Programs in the tests directory are now build with target tests and placed in the same location. * clean target was expanded to remove new binaries * test target binaries are listed in a variable * Locations of binaries were added to the .gitignore Signed-off-by: Jiri Podivin Co-authored-by: Georgi Gerganov --- .gitignore | 9 +++++++++ Makefile | 30 ++++++++++++++++++++++++++---- 2 files changed, 35 insertions(+), 4 deletions(-) (limited to '.gitignore') diff --git a/.gitignore b/.gitignore index a23ac59..9193930 100644 --- a/.gitignore +++ b/.gitignore @@ -61,3 +61,12 @@ qnt-*.txt perf-*.txt examples/jeopardy/results.txt + +# Test binaries +tests/test-double-float +tests/test-grad0 +tests/test-opt +tests/test-quantize-fns +tests/test-quantize-perf +tests/test-sampling +tests/test-tokenizer-0 diff --git a/Makefile b/Makefile index cff4d97..61f2c77 100644 --- a/Makefile +++ b/Makefile @@ -1,6 +1,9 @@ # Define the default target now so that it is always the first target BUILD_TARGETS = main quantize quantize-stats perplexity embedding vdot train-text-from-scratch simple server embd-input-test +# Binaries only useful for tests +TEST_TARGETS = tests/test-double-float tests/test-grad0 tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling tests/test-tokenizer-0 + default: $(BUILD_TARGETS) ifndef UNAME_S @@ -316,7 +319,7 @@ libllama.so: llama.o ggml.o $(OBJS) $(CXX) $(CXXFLAGS) -shared -fPIC -o $@ $^ $(LDFLAGS) clean: - rm -vf *.o *.so *.dll main quantize quantize-stats perplexity embedding benchmark-matmult save-load-state server simple vdot train-text-from-scratch embd-input-test build-info.h + rm -vf *.o *.so *.dll main quantize quantize-stats perplexity embedding benchmark-matmult save-load-state server simple vdot train-text-from-scratch embd-input-test build-info.h $(TEST_TARGETS) # # Examples @@ -371,6 +374,8 @@ build-info.h: $(wildcard .git/index) scripts/build-info.sh # Tests # +tests: $(TEST_TARGETS) + benchmark-matmult: examples/benchmark/benchmark-matmult.cpp build-info.h ggml.o $(OBJS) $(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS) ./$@ @@ -378,6 +383,23 @@ benchmark-matmult: examples/benchmark/benchmark-matmult.cpp build-info.h ggml.o vdot: pocs/vdot/vdot.cpp ggml.o $(OBJS) $(CXX) $(CXXFLAGS) $^ -o $@ $(LDFLAGS) -.PHONY: tests clean -tests: - bash ./tests/run-tests.sh +tests/test-double-float: tests/test-double-float.c build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) + +tests/test-grad0: tests/test-grad0.c build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) + +tests/test-opt: tests/test-opt.c build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) + +tests/test-quantize-fns: tests/test-quantize-fns.cpp build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) + +tests/test-quantize-perf: tests/test-quantize-perf.cpp build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) + +tests/test-sampling: tests/test-sampling.cpp build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) + +tests/test-tokenizer-0: tests/test-tokenizer-0.cpp build-info.h ggml.o llama.o common.o $(OBJS) + $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) -- cgit v1.2.3 From 73643f5fb1136dc2b65ae910bdc5a431520d70a2 Mon Sep 17 00:00:00 2001 From: Jose Maldonado <63384398+yukiteruamano@users.noreply.github.com> Date: Fri, 21 Jul 2023 06:53:27 -0400 Subject: gitignore : changes for Poetry users + chat examples (#2284) A fix in Makefile for FreeBSD users. In the platfrom x86_64 is amd64. This fix resolve compilation using CFLAGS and CXXFLAGS with -march=native and -mtune=native Add two examples for interactive mode using Llama2 models (thx TheBloke for models) Co-authored-by: Georgi Gerganov --- .gitignore | 7 ++++++- Makefile | 2 +- examples/llama2-13b.sh | 18 ++++++++++++++++++ examples/llama2.sh | 18 ++++++++++++++++++ 4 files changed, 43 insertions(+), 2 deletions(-) create mode 100755 examples/llama2-13b.sh create mode 100755 examples/llama2.sh (limited to '.gitignore') diff --git a/.gitignore b/.gitignore index 9193930..c26d82a 100644 --- a/.gitignore +++ b/.gitignore @@ -62,6 +62,11 @@ perf-*.txt examples/jeopardy/results.txt + +pyproject.toml +poetry.lock +poetry.toml + # Test binaries tests/test-double-float tests/test-grad0 @@ -69,4 +74,4 @@ tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling -tests/test-tokenizer-0 +tests/test-tokenizer-0 \ No newline at end of file diff --git a/Makefile b/Makefile index 4f8c4b3..1ea3c45 100644 --- a/Makefile +++ b/Makefile @@ -127,7 +127,7 @@ endif # Architecture specific # TODO: probably these flags need to be tweaked on some architectures # feel free to update the Makefile for your architecture and send a pull request or issue -ifeq ($(UNAME_M),$(filter $(UNAME_M),x86_64 i686)) +ifeq ($(UNAME_M),$(filter $(UNAME_M),x86_64 i686 amd64)) # Use all CPU extensions that are available: CFLAGS += -march=native -mtune=native CXXFLAGS += -march=native -mtune=native diff --git a/examples/llama2-13b.sh b/examples/llama2-13b.sh new file mode 100755 index 0000000..92b3f6d --- /dev/null +++ b/examples/llama2-13b.sh @@ -0,0 +1,18 @@ +#!/bin/bash + +# +# Temporary script - will be removed in the future +# + +cd `dirname $0` +cd .. + +./main -m models/available/Llama2/13B/llama-2-13b.ggmlv3.q4_0.bin \ + --color \ + --ctx_size 2048 \ + -n -1 \ + -ins -b 256 \ + --top_k 10000 \ + --temp 0.2 \ + --repeat_penalty 1.1 \ + -t 8 diff --git a/examples/llama2.sh b/examples/llama2.sh new file mode 100755 index 0000000..221b375 --- /dev/null +++ b/examples/llama2.sh @@ -0,0 +1,18 @@ +#!/bin/bash + +# +# Temporary script - will be removed in the future +# + +cd `dirname $0` +cd .. + +./main -m models/available/Llama2/7B/llama-2-7b.ggmlv3.q4_0.bin \ + --color \ + --ctx_size 2048 \ + -n -1 \ + -ins -b 256 \ + --top_k 10000 \ + --temp 0.2 \ + --repeat_penalty 1.1 \ + -t 8 -- cgit v1.2.3 From 3973b25a64a37a47eac156a3fd28f83c16f14bf2 Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Fri, 21 Jul 2023 14:42:41 +0300 Subject: gitignore : fix final newline --- .gitignore | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to '.gitignore') diff --git a/.gitignore b/.gitignore index c26d82a..c1ab6bb 100644 --- a/.gitignore +++ b/.gitignore @@ -74,4 +74,5 @@ tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling -tests/test-tokenizer-0 \ No newline at end of file +tests/test-tokenizer-0 + -- cgit v1.2.3