def get_version(flags):
device_id = get_cur_device_id(flags)
if device_id is not None:
android_device = device.create_device(flags.target_abi, device_id)
target_props = android_device.info()
target_soc = target_props["ro.board.platform"]
android_ver = (int)(target_props["ro.build.version.release"])
apu_version = get_apu_version(True, android_ver, target_soc)
else:
apu_version = 4
MaceLogger.warning("Can not get unique device ID, MACE select the"
" latest apu version: %s" % apu_version)
sys.exit(apu_version)
def get_cur_device_id(flags):
run_devices = device.choose_devices(flags.target_abi, "all")
run_device = None
device_num = len(run_devices)
if device_num == 0: # for CI
MaceLogger.warning("No Android devices are plugged in, "
"you need to copy `apu` so files by yourself.")
elif device_num > 1: # for CI
MaceLogger.warning("More than one Android devices are plugged in, "
"you need to copy `apu` so files by yourself.")
else:
run_device = run_devices[0]
return run_device
def merge_opencl_binaries(opencl_binaries, output_file):
platform_info_key = 'mace_opencl_precompiled_platform_info_key'
kvs = {}
for binary in opencl_binaries:
if not os.path.exists(binary):
MaceLogger.warning("OpenCL bin %s not found" % binary)
continue
with open(binary, "rb") as f:
binary_array = np.fromfile(f, dtype=np.uint8)
idx = 0
size, = struct.unpack("Q", binary_array[idx:idx + 8])
idx += 8
for _ in range(size):
key_size, = struct.unpack("i", binary_array[idx:idx + 4])
idx += 4
key, = struct.unpack(
str(key_size) + "s", binary_array[idx:idx + key_size])
idx += key_size
value_size, = struct.unpack("i", binary_array[idx:idx + 4])
idx += 4
if key == platform_info_key and key in kvs:
mace_check(
(kvs[key] == binary_array[idx:idx + value_size]).all(),
"There exists more than one OpenCL version for models:"
" %s vs %s " %
(kvs[key], binary_array[idx:idx + value_size]))
else:
kvs[key] = binary_array[idx:idx + value_size]
idx += value_size
output_byte_array = bytearray()
data_size = len(kvs)
output_byte_array.extend(struct.pack("Q", data_size))
for key, value in kvs.items():
key_size = len(key)
output_byte_array.extend(struct.pack("i", key_size))
output_byte_array.extend(struct.pack(str(key_size) + "s", key))
value_size = len(value)
output_byte_array.extend(struct.pack("i", value_size))
output_byte_array.extend(value)
np.array(output_byte_array).tofile(output_file)
def merge_opencl_parameters(params_files, output_file):
kvs = {}
for params in params_files:
if not os.path.exists(params):
MaceLogger.warning("Tune param %s not found" % params)
continue
with open(params, "rb") as f:
binary_array = np.fromfile(f, dtype=np.uint8)
idx = 0
size, = struct.unpack("Q", binary_array[idx:idx + 8])
idx += 8
for _ in range(size):
key_size, = struct.unpack("i", binary_array[idx:idx + 4])
idx += 4
key, = struct.unpack(
str(key_size) + "s", binary_array[idx:idx + key_size])
idx += key_size
value_size, = struct.unpack("i", binary_array[idx:idx + 4])
idx += 4
kvs[key] = binary_array[idx:idx + value_size]
idx += value_size
output_byte_array = bytearray()
data_size = len(kvs)
output_byte_array.extend(struct.pack("Q", data_size))
for key, value in kvs.items():
key_size = len(key)
output_byte_array.extend(struct.pack("i", key_size))
output_byte_array.extend(struct.pack(str(key_size) + "s", key))
value_size = len(value)
output_byte_array.extend(struct.pack("i", value_size))
output_byte_array.extend(value)
np.array(output_byte_array).tofile(output_file)
def run_model_for_device(flags, args, dev, model_name, model_conf):
target_abi = flags.target_abi
install_dir = run_target.default_install_dir(target_abi) + "/" + model_name
sysdir = install_dir + "/interior"
dev.mkdir(sysdir)
runtime_list = []
for graph_name, graph_conf in model_conf[ModelKeys.subgraphs].items():
runtime = graph_conf[ModelKeys.runtime]
runtime_list.append(runtime)
mace_check(runtime != DeviceType.APU or target_abi == "arm64-v8a",
"APU runtime does only support arm64-v8a")
# install models to devices
workdir = flags.output + "/" + model_name
model_file = model_name + ".pb"
model_data_file = model_name + ".data"
model_path = workdir + "/model/" + model_file
model_data_path = workdir + "/model/" + model_data_file
if os.path.exists(model_path) and os.path.exists(model_data_path):
dev.install(Target(model_path), install_dir)
dev.install(Target(model_data_path), install_dir)
else:
MaceLogger.warning("No models exist in %s, use --model_file and"
" --model_data_file specified in args" % model_path)
if ModelKeys.check_tensors in model_conf:
model_conf[ModelKeys.output_tensors] = model_conf[
ModelKeys.check_tensors]
model_conf[ModelKeys.output_shapes] = model_conf[
ModelKeys.check_shapes]
model_file_path = ""
if not flags.gencode_model:
model_file_path = install_dir + "/" + model_file
model_data_file_path = ""
if not flags.gencode_param:
model_data_file_path = install_dir + "/" + model_data_file
input_tensors_info = config_parser.find_input_tensors_info(
model_conf[ModelKeys.subgraphs], model_conf[ModelKeys.input_tensors])
output_tensors_info = config_parser.find_output_tensors_info(
model_conf[ModelKeys.subgraphs], model_conf[ModelKeys.output_tensors])
model_args = {
"model_name":
model_name,
"model_file":
model_file_path,
"model_data_file":
model_data_file_path,
"input_node":
",".join(model_conf[ModelKeys.input_tensors]),
"input_shape":
join_2d_array(input_tensors_info[ModelKeys.input_shapes]),
"output_node":
",".join(model_conf[ModelKeys.output_tensors]),
"output_shape":
join_2d_array(output_tensors_info[ModelKeys.output_shapes]),
"input_data_format":
",".join([
df.name for df in input_tensors_info[ModelKeys.input_data_formats]
]),
"output_data_format":
",".join([
df.name
for df in output_tensors_info[ModelKeys.output_data_formats]
])
}
opts = [
"--%s='%s'" % (arg_key, arg_val)
for arg_key, arg_val in model_args.items()
] + args
should_generate_data = (flags.validate or flags.tune
or "--benchmark" in opts)
if should_generate_data:
tmpdirname = tempfile.mkdtemp()
input_file_prefix = tmpdirname + "/" + model_name
if ModelKeys.validation_inputs_data in model_conf:
input_tensor = model_conf[ModelKeys.input_tensors]
input_data = model_conf[ModelKeys.validation_inputs_data]
mace_check(
len(input_tensor) == len(input_data),
"len(input_tensor) != len(validate_data")
for i in range(len(input_tensor)):
util.download_or_get_file(
model_conf[ModelKeys.validation_inputs_data][i], "",
util.formatted_file_name(input_file_prefix,
input_tensor[i]))
else:
generate_input_data(input_file_prefix,
model_conf[ModelKeys.input_tensors],
input_tensors_info[ModelKeys.input_shapes],
input_tensors_info[ModelKeys.input_ranges],
input_tensors_info[ModelKeys.input_data_types])
dev.install(Target(tmpdirname), install_dir + "/validate_in")
target_input_file = "%s/validate_in/%s" % (install_dir, model_name)
target_output_dir = "%s/validate_out" % install_dir
dev.mkdir(target_output_dir)
target_output_file = target_output_dir + "/" + model_name
opts += [
"--input_file=%s" % target_input_file,
"--output_file=%s" % target_output_file
]
# run
envs = flags.envs.split(" ") + ["MACE_INTERNAL_STORAGE_PATH=%s" % sysdir]
if flags.tune:
envs += [
"MACE_TUNING=1",
"MACE_RUN_PARAMETER_PATH=%s/interior/tune_params" % install_dir
]
opts += ["--round=0"]
mace_check(flags.vlog_level >= 0,
"vlog_level should be greater than zeror")
envs += ["MACE_CPP_MIN_VLOG_LEVEL=%s" % flags.vlog_level]
build_dir = flags.build_dir + "/" + target_abi
libs = []
if DeviceType.HEXAGON in runtime_list:
libs += ["third_party/nnlib/%s/libhexagon_controller.so" % target_abi]
elif runtime == DeviceType.HTA:
libs += ["third_party/hta/%s/libhta_hexagon_runtime.so" % target_abi]
elif DeviceType.APU in runtime_list:
apu_libs = get_apu_so_paths(dev)
libs += apu_libs
cpp_shared_lib_path = os.path.join(build_dir,
"install/lib/libc++_shared.so")
if os.path.exists(cpp_shared_lib_path):
libs.append(cpp_shared_lib_path)
target = Target(build_dir + "/install/bin/mace_run",
libs,
opts=opts,
envs=envs)
run_target.run_target(target_abi, install_dir, target, dev)
if DeviceType.GPU in runtime_list:
opencl_dir = workdir + "/opencl"
util.mkdir_p(opencl_dir)
dev.pull(
Target(install_dir + "/interior/mace_cl_compiled_program.bin"),
"%s/%s_compiled_opencl_kernel.%s.%s.bin" %
(opencl_dir, model_name, dev.info()["ro.product.model"].replace(
' ', ''), dev.info()["ro.board.platform"]))
if flags.tune:
dev.pull(
Target(install_dir + "/interior/tune_params"),
"%s/%s_tuned_opencl_parameter.%s.%s.bin" %
(opencl_dir, model_name,
dev.info()["ro.product.model"].replace(
' ', ''), dev.info()["ro.board.platform"]))
if flags.validate:
validate_model_file = util.download_or_get_model(
model_conf[ModelKeys.model_file_path],
model_conf[ModelKeys.model_sha256_checksum], tmpdirname)
validate_weight_file = ""
if ModelKeys.weight_file_path in model_conf:
validate_weight_file = util.download_or_get_model(
model_conf[ModelKeys.weight_file_path],
model_conf[ModelKeys.weight_sha256_checksum], tmpdirname)
dev.pull(Target(target_output_dir), tmpdirname + "/validate_out")
output_file_prefix = tmpdirname + "/validate_out/" + model_name
validate.validate(model_conf[ModelKeys.platform], validate_model_file,
validate_weight_file, input_file_prefix,
output_file_prefix,
input_tensors_info[ModelKeys.input_shapes],
output_tensors_info[ModelKeys.output_shapes],
input_tensors_info[ModelKeys.input_data_formats],
output_tensors_info[ModelKeys.output_data_formats],
input_tensors_info[ModelKeys.input_tensors],
output_tensors_info[ModelKeys.output_tensors],
flags.validate_threshold,
input_tensors_info[ModelKeys.input_data_types],
flags.backend, "", "")
if should_generate_data:
shutil.rmtree(tmpdirname)
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
from utils import device
from utils.util import MaceLogger
cwd = os.path.dirname(__file__)
# TODO: Remove bazel deps
try:
device.execute("bazel version")
except: # noqa
MaceLogger.warning("No bazel, use cmake.")
device.execute("bash tools/cmake/cmake-generate-proto-py-host.sh")
else:
try:
device.execute("bazel build //mace/proto:mace_py")
device.execute("cp -f bazel-genfiles/mace/proto/mace_pb2.py %s" % cwd)
device.execute("bazel build //mace/proto:micro_mem_py")
device.execute("cp -f bazel-genfiles/mace/proto/micro_mem_pb2.py %s" %
cwd)
device.execute("bazel build //third_party/caffe:caffe_py")
device.execute(
"cp -f bazel-genfiles/third_party/caffe/caffe_pb2.py %s" % cwd)
except: # noqa
MaceLogger.error("Failed in proto files' building")
default="arm64-v8a",
help="Target ABI: only support arm64-v8a")
parser.add_argument("--target_soc",
type=str,
default="all",
help="serialno for adb connection")
parser.add_argument("--apu_path",
type=str,
default="",
help="path for storing apu so files on device")
return parser.parse_known_args()
if __name__ == "__main__":
flags, args = parse_args()
run_devices = device.choose_devices(flags.target_abi, flags.target_soc)
device_num = len(run_devices)
if device_num == 0: # for CI
MaceLogger.warning("No Android devices are plugged in, "
"you need to copy `apu` so files by yourself.")
elif device_num > 1: # for CI
MaceLogger.warning("More than one Android devices are plugged in, "
"you need to copy `apu` so files by yourself.")
else:
device_id = run_devices[0]
android_device = device.create_device(flags.target_abi, device_id)
apu_so_paths = get_apu_so_paths(android_device)
for apu_so_path in apu_so_paths:
device.execute("cp -f %s %s" % (apu_so_path, flags.apu_path), True)