Build Tensorflow Lite on x86_64, Raspberry Pi and other ARM

1. git clone https://github.com/tensorflow/tensorflow.git

2.
cd tensorflow/contrib/lite

3. Download dependencies and MobileNet models
	./download_dependencies.sh

4.
# Before building, you need to enable SIMD instruction sets for your processor for build to succeed.
# Run 'cat /proc/cpuinfo'. If you see sse4* or neon* in flags, you have SIMD capable CPU.
# Edit tensorflow/contrib/lite/Makefile, and add one of these to CXXFLAGS and CFLAGS:
# 	"-msse4" for x86_64 CPU
# 	"-mfpu=neon" for ARM CPU. For Raspberry Pi 2, add "-mfpu=neon-vfpv4".
#
# If you don't see any SIMD instruction set, add this define instead to CXXFLAGS and CFLAGS:
# 	-DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK

5. Now make TFLite:
	cd ../../..
	make -f tensorflow/contrib/lite/Makefile

# This generates tensorflow/contrib/lite/gen/lib/libtensorflow-lite.a static library.

6.
# See tensorflow/contrib/lite/examples/label_image/label_image.cc to understand how to use
# TFLite API.

7.
# To build the label_image example app:
# Create the following Makefile as tensorflow/contrib/lite/examples/label_image/Makefile

########### START OF LABEL_IMAGE MAKEFILE ####################################################
# Find where we're running from, so we can store generated files here.
ifeq ($(origin MAKEFILE_DIR), undefined)
	MAKEFILE_DIR := $(shell dirname $(realpath $(lastword $(MAKEFILE_LIST))))
endif

# Try to figure out the host system
HOST_OS :=
ifeq ($(OS),Windows_NT)
	HOST_OS = WINDOWS
else
	UNAME_S := $(shell uname -s)
	ifeq ($(UNAME_S),Linux)
	        HOST_OS := LINUX
	endif
	ifeq ($(UNAME_S),Darwin)
		HOST_OS := OSX
	endif
endif

ARCH := $(shell if [[ $(shell uname -m) =~ i[345678]86 ]]; then echo x86_32; else echo $(shell uname -m); fi)

# Where compiled objects are stored.
OBJDIR := $(MAKEFILE_DIR)/gen/obj/
BINDIR := $(MAKEFILE_DIR)/gen/bin/
LIBDIR := $(MAKEFILE_DIR)/gen/lib/
GENDIR := $(MAKEFILE_DIR)/gen/obj/

# Settings for the host compiler.
CXX := $(CC_PREFIX) gcc
CXXFLAGS := --std=c++11 -O3 -DNDEBUG
CC := $(CC_PREFIX) gcc
CFLAGS :=
LDOPTS :=
LDOPTS += -L/usr/local/lib
ARFLAGS := -r

INCLUDES := \
-I. \
-I$(MAKEFILE_DIR)/../../../../../ \
-I$(MAKEFILE_DIR)/../../downloads/flatbuffers/include \
-I$(GENDIR)
# This is at the end so any globally-installed frameworks like protobuf don't
# override local versions in the source tree.
INCLUDES += -I/usr/local/include

LIBS := \
-lstdc++ \
-lpthread \
-lm \
-lz

# If we're on Linux, also link in the dl library.
ifeq ($(HOST_OS),LINUX)
	LIBS += -ldl -lpthread
endif

# This library is the main target for this makefile. It will contain a minimal
# runtime that can be linked in to other programs.
BIN_PATH := $(BINDIR)label_image

TF_LIB_PATH := $(MAKEFILE_DIR)/../../gen/lib/libtensorflow-lite.a

# What sources we want to compile, must be kept in sync with the main Bazel
# build files.

SRCS := \
tensorflow/contrib/lite/examples/label_image/label_image.cc \
tensorflow/contrib/lite/examples/label_image/bitmap_helpers.cc

OBJS := $(addprefix $(OBJDIR), \
$(patsubst %.cc,%.o,$(patsubst %.c,%.o,$(SRCS))))

# For normal manually-created TensorFlow C++ source files.
$(OBJDIR)%.o: %.cc
	@mkdir -p $(dir $@)
	$(CXX) $(CXXFLAGS) $(INCLUDES) -c $< -o $@

# For normal manually-created TensorFlow C++ source files.
$(OBJDIR)%.o: %.c
	@mkdir -p $(dir $@)
	$(CC) $(CCFLAGS) $(INCLUDES) -c $< -o $@

# The target that's compiled if there's no command-line arguments.
all: $(BIN_PATH)

$(BIN_PATH): $(OBJS) $(TF_LIB_PATH)
	@mkdir -p $(dir $@)
	$(CXX) $(CXXFLAGS) $(INCLUDES) \
	-o $(BIN_PATH) $(OBJS) \
	$(LIBFLAGS) $(TF_LIB_PATH) $(LDFLAGS) $(LIBS)

# Gets rid of all generated files.
clean:
	rm -rf $(MAKEFILE_DIR)/gen

# Gets rid of target files only, leaving the host alone. Also leaves the lib
# directory untouched deliberately, so we can persist multiple architectures
# across builds for iOS and Android.
cleantarget:
	rm -rf $(OBJDIR)
	rm -rf $(BINDIR)

$(DEPDIR)/%.d: ;
.PRECIOUS: $(DEPDIR)/%.d

-include $(patsubst %,$(DEPDIR)/%.d,$(basename $(TF_CC_SRCS)))




########### END OF LABEL_IMAGE MAKEFILE ####################################################



8. Make the label_image app:
	make -f tensorflow/contrib/lite/examples/label_image/Makefile

# The app binary is generated in: tensorflow/contrib/lite/examples/label_image/gen/bin/label_image


9. Run label_image inference using floating point model:
	cd  tensorflow/contrib/lite/examples/label_image/gen/bin/

	./label_image -v 1 -m ../../../../downloads/models/models/mobilenet_v1_1.0_224.tflite  -i ../../testdata/grace_hopper.bmp -l ../../../../downloads/models/models/labels.txt


10. Run label_image inference using quantized model:
	cd  tensorflow/contrib/lite/examples/label_image/gen/bin/

	./label_image -v 1 -m ../../../../downloads/quantized_models/mobilenet_quant_v1_224.tflite  -i ../../testdata/grace_hopper.bmp -l ../../../../downloads/quantized_models/labels.txt