chore: Make help (#25)

Add `help` target to Makefile which extracts `##` help-comments
from the Makefile itself, and add help text for common targets.

Author: ckadner

Makefile

@@ -5,73 +5,99 @@ TEST_IMAGE_NAME ?= 'cpu-tests:0'
 SERVER_IMAGE_NAME ?= 'text-gen-server:0'
 GIT_COMMIT_HASH := $(shell git rev-parse --short HEAD)
 
-build:
+.PHONY: all
+all: help
+
+.PHONY: help
+help: ## Display this help
+	@awk 'BEGIN{FS=":.*##"; printf("\nUsage:\n  make \033[36m<target>\033[0m\n\nTargets:\n")} /^[-a-zA-Z_0-9\\.]+:.*?##/ {t=$$1; if(!(t in p)){p[t]; printf("  \033[36m%-20s\033[0m %s\n", t, $$2)}}' $(MAKEFILE_LIST)
+
+.PHONY: build
+build: ## Build server release image
 	docker build --progress=plain --target=server-release --build-arg GIT_COMMIT_HASH=$(GIT_COMMIT_HASH) -t $(SERVER_IMAGE_NAME) .
 	docker images
 
-all: help
-
+.PHONY: install-server
 install-server:
 	cd server && make install
 
+.PHONY: install-custom-kernels
 install-custom-kernels:
-	if [ "$$BUILD_EXTENSIONS" = "True" ]; then cd server/custom_kernels && python setup.py install; else echo "Custom kernels are disabled, you need to set the BUILD_EXTENSIONS environment variable to 'True' in order to build them. (Please read the docs, kernels might not work on all hardware)"; fi
-
+	if [ "$$BUILD_EXTENSIONS" = "True" ]; then \
+		cd server/custom_kernels && \
+		python setup.py install; \
+	else \
+		echo "Custom kernels are disabled, you need to set the BUILD_EXTENSIONS environment variable to 'True' in order to build them. (Please read the docs, kernels might not work on all hardware)"; \
+	fi
+
+.PHONY: install-router
 install-router:
 	cd router && cargo install --path .
 
+.PHONY: install-launcher
 install-launcher:
 	cd launcher && env GIT_COMMIT_HASH=$(GIT_COMMIT_HASH) cargo install --path .
 
 .PHONY: install-launcher-linux
 install-launcher-linux:
 	cd launcher && env GIT_COMMIT_HASH=$(GIT_COMMIT_HASH) CARGO_TARGET_X86_64_UNKNOWN_LINUX_GNU_LINKER=x86_64-unknown-linux-gnu-gcc cargo install --path . --target=x86_64-unknown-linux-gnu
 
-install: install-server install-router install-launcher install-custom-kernels
+.PHONY: install
+install: install-server install-router install-launcher install-custom-kernels ## Install server, router, launcher, and custom kernels
 
+.PHONY: server-dev
 server-dev:
 	cd server && make run-dev
 
+.PHONY: router-dev
 router-dev:
 	cd router && cargo run
 
+.PHONY: run-bloom-560m
 run-bloom-560m:
 	text-generation-launcher --model-name bigscience/bloom-560m --num-shard 2
 
+.PHONY: run-bloom-560m-quantize
 run-bloom-560m-quantize:
 	text-generation-launcher --model-name bigscience/bloom-560m --num-shard 2 --dtype-str int8
 
+.PHONY: download-bloom
 download-bloom:
 	text-generation-server download-weights bigscience/bloom
 
+.PHONY: run-bloom
 run-bloom:
 	text-generation-launcher --model-name bigscience/bloom --num-shard 8
 
+.PHONY: run-bloom-quantize
 run-bloom-quantize:
 	text-generation-launcher --model-name bigscience/bloom --num-shard 8 --dtype-str int8
 
-build-test-image:
+.PHONY: build-test-image
+build-test-image: ## Build the test image
 	docker build --progress=plain --target=cpu-tests -t $(TEST_IMAGE_NAME) .
 
+.PHONY: check-test-image
 check-test-image:
 	@docker image inspect $(TEST_IMAGE_NAME) >/dev/null 2>&1 || $(MAKE) build-test-image
 
-integration-tests: check-test-image
+.PHONY: integration-tests
+integration-tests: check-test-image ## Run integration tests
 	mkdir -p /tmp/transformers_cache
 	docker run --rm -v /tmp/transformers_cache:/transformers_cache \
 		-e HUGGINGFACE_HUB_CACHE=/transformers_cache \
 		-e TRANSFORMERS_CACHE=/transformers_cache \
 		-w /usr/src/integration_tests \
 		$(TEST_IMAGE_NAME) make test
 
-python-tests: check-test-image
+.PHONY: python-tests
+python-tests: check-test-image ## Run Python tests
 	mkdir -p /tmp/transformers_cache
 	docker run --rm -v /tmp/transformers_cache:/transformers_cache \
 		-e HUGGINGFACE_HUB_CACHE=/transformers_cache \
 		-e TRANSFORMERS_CACHE=/transformers_cache \
 		$(TEST_IMAGE_NAME) pytest -sv --ignore=server/tests/test_utils.py server/tests
 
+.PHONY: clean
 clean:
 	rm -rf target
-
-.PHONY: build build-test-image integration-tests python-tests

server/Makefile

@@ -1,26 +1,38 @@
+.PHONY: all
+all: install run-dev
+
+.PHONY: gen-server
 gen-server:
 	# Compile protos
 	pip install grpcio-tools==1.60.0 mypy-protobuf==3.5.0 'types-protobuf>=3.20.4' --no-cache-dir
-	mkdir text_generation_server/pb || true
-	python -m grpc_tools.protoc -I../proto --python_out=text_generation_server/pb \
-		--grpc_python_out=text_generation_server/pb --mypy_out=text_generation_server/pb ../proto/generate.proto
+	mkdir -p text_generation_server/pb
+	python -m grpc_tools.protoc -I../proto \
+		--python_out=text_generation_server/pb \
+		--grpc_python_out=text_generation_server/pb \
+		--mypy_out=text_generation_server/pb \
+		../proto/generate.proto
 	find text_generation_server/pb/ -type f -name "*.py" -print0 -exec sed -i -e 's/^\(import.*pb2\)/from . \1/g' {} \;
 	touch text_generation_server/pb/__init__.py
 
 TORCH_URL := https://download.pytorch.org/whl/cu118
 TORCH_VERSION := 2.0.0+cu118
 
+.PHONY: install-torch
 install-torch:
 	# Install specific version of torch
 	pip install ninja==1.11.1.1 torch==$(TORCH_VERSION) --extra-index-url $(TORCH_URL) --no-cache-dir
 
+.PHONY: install-deepspeed
 install-deepspeed:
 	# Install specific version of deepspeed
 	pip install deepspeed==0.7.6 --no-cache-dir
 
+.PHONY: install
 install: gen-server install-torch
 	pip install pip --upgrade
 	pip install -e . --no-cache-dir
 
+.PHONY: run-dev
 run-dev:
-	SAFETENSORS_FAST_GPU=1 python -m torch.distributed.run --nproc_per_node=2 text_generation_server/cli.py serve bigscience/bloom-560m
+	SAFETENSORS_FAST_GPU=1 python -m torch.distributed.run --nproc_per_node=2 \
+		text_generation_server/cli.py serve bigscience/bloom-560m hf_transformers