def _evaluate_single_model(self, record):
try:
worker_info = {
"step_name": record.step_name,
"worker_id": record.worker_id
}
_record = dict(worker_id=record.worker_id,
desc=record.desc,
step_name=record.step_name)
_init_record = ReportRecord().load_dict(_record)
Report().broadcast(_init_record)
if EvaluatorConfig.gpu_evaluator_enable:
cls_evaluator = ClassFactory.get_cls(ClassType.GPU_EVALUATOR,
"GpuEvaluator")
evaluator = cls_evaluator(worker_info=worker_info,
model_desc=record.desc,
weights_file=record.weights_file)
self.master.run(evaluator)
if EvaluatorConfig.davinci_mobile_evaluator_enable:
cls_evaluator = ClassFactory.get_cls(
ClassType.DAVINCI_MOBILE_EVALUATOR,
"DavinciMobileEvaluator")
evaluator = cls_evaluator(worker_info=worker_info,
model_desc=record.desc,
weights_file=record.weights_file)
self.master.run(evaluator)
except Exception:
logger.error(
"Failed to evaluate model, worker info={}".format(worker_info))
logger.error(traceback.format_exc())
return
def from_desc(cls, desc):
"""Create Model from desc."""
desc = deepcopy(desc)
module_groups = desc.get('modules', [])
module_type = desc.get('type', 'Sequential')
loss = desc.get('loss')
modules = OrderedDict()
for group_name in module_groups:
module_desc = deepcopy(desc.get(group_name))
if 'modules' in module_desc:
module = cls.from_desc(module_desc)
else:
cls_name = module_desc.get('type')
if not ClassFactory.is_exists(ClassType.NETWORK, cls_name):
raise ValueError("Network {} not exists.".format(cls_name))
module = ClassFactory.get_instance(ClassType.NETWORK,
module_desc)
modules[group_name] = module
if not modules and module_type:
model = ClassFactory.get_instance(ClassType.NETWORK, desc)
else:
if ClassFactory.is_exists(SearchSpaceType.CONNECTIONS,
module_type):
connections = ClassFactory.get_cls(SearchSpaceType.CONNECTIONS,
module_type)
else:
connections = ClassFactory.get_cls(SearchSpaceType.CONNECTIONS,
'Sequential')
model = list(modules.values())[0] if len(
modules) == 1 else connections(modules)
if loss:
model.add_loss(ClassFactory.get_cls(ClassType.LOSS, loss))
return model
def from_desc(cls, desc):
"""Create Model from desc."""
module_groups = desc.get('modules')
module_type = desc.get('type', 'Sequential')
loss = desc.get('loss')
modules = OrderedDict()
for group_name in module_groups:
module_desc = deepcopy(desc.get(group_name))
if 'modules' in module_desc:
module = cls.from_desc(module_desc)
else:
cls_name = module_desc.get('type')
if not ClassFactory.is_exists(ClassType.SEARCH_SPACE,
cls_name):
return None
module = ClassFactory.get_instance(ClassType.SEARCH_SPACE,
module_desc)
modules[group_name] = module
if ClassFactory.is_exists(SearchSpaceType.CONNECTIONS, module_type):
connections = ClassFactory.get_cls(SearchSpaceType.CONNECTIONS,
module_type)
else:
connections = ClassFactory.get_cls(SearchSpaceType.CONNECTIONS,
'Sequential')
model = list(
modules.values())[0] if len(modules) == 1 else connections(modules)
if loss:
model.add_loss(ClassFactory.get_cls(ClassType.LOSS, loss))
return model
def __init__(self, generator, discriminator, latent_dim, gen_bs):
super(GAN, self).__init__()
self.generator = ClassFactory.get_cls(
ClassType.NETWORK, generator.pop('type'))(**generator)
self.latent_dim = latent_dim
self.gen_bs = gen_bs
self.discriminator = ClassFactory.get_cls(
ClassType.NETWORK, discriminator.pop('type'))(**discriminator)
def __new__(cls, *args, **kwargs):
"""Create search algorithm instance by ClassFactory."""
if cls.__name__ != 'Codec':
return super().__new__(cls)
if kwargs.get('type'):
t_cls = ClassFactory.get_cls(ClassType.CODEC, kwargs.pop('type'))
else:
t_cls = ClassFactory.get_cls(ClassType.CODEC)
return super().__new__(t_cls)
def __new__(cls, *args, **kwargs):
"""Create a subclass instance of dataset."""
if Dataset in cls.__bases__:
return super().__new__(cls)
if kwargs.get('type'):
t_cls = ClassFactory.get_cls(ClassType.DATASET, kwargs.pop('type'))
else:
t_cls = ClassFactory.get_cls(ClassType.DATASET)
return super().__new__(t_cls)
def __new__(cls, *args, **kwargs):
"""Create search algorithm instance by ClassFactory."""
if cls.__name__ != 'SearchAlgorithm':
return super().__new__(cls)
if kwargs.get('type'):
t_cls = ClassFactory.get_cls(ClassType.SEARCH_ALGORITHM, kwargs.pop('type'))
else:
t_cls = ClassFactory.get_cls(ClassType.SEARCH_ALGORITHM, PipeStepConfig.search_algorithm.type)
return super().__new__(t_cls)
def __init__(self, type):
if type not in ['restrict', 'target']:
raise ValueError('Input type must be restriction or target.')
self.filter_types, self.terminate_types = [], []
self.filter_compares = dict()
self.terminate_compares = dict()
self.filters_to_params = dict()
self._init_compare_types(type)
for filter in self.filter_types:
t_cls = ClassFactory.get_cls(ClassType.QUOTA, filter)
self.filter_compares[filter] = t_cls()
for terminate in self.terminate_types:
t_cls = ClassFactory.get_cls(ClassType.QUOTA, terminate)
self.terminate_compares[terminate] = t_cls()
self.filter_rules = copy.deepcopy(General.quota.filter_rules)
def _init_dataloader(self, mode, loader=None):
"""Init dataloader."""
if loader is not None:
return loader
if mode == "train" and self.hps is not None and self.hps.get("dataset") is not None:
dataset_cls = ClassFactory.get_cls(ClassType.DATASET)
dataset = dataset_cls(mode=mode, hps=self.hps.get("dataset"))
else:
dataset_cls = ClassFactory.get_cls(ClassType.DATASET)
dataset = dataset_cls(mode=mode)
if self.distributed and mode == "train":
dataset.set_distributed(self._world_size, self._rank_id)
# adapt the dataset to specific backend
dataloader = Adapter(dataset).loader
return dataloader
def __init__(self,
num_classes,
backbone='SerialBackbone',
neck='TorchFPN',
network_name='torchvision_FasterRCNN',
weight_file=None,
**kwargs):
"""Create layers.
:param num_class: number of class
:type num_class: int
"""
super(FasterRCNN, self).__init__()
self.weight_file = weight_file
backbone_cls = self.define_props('backbone',
backbone,
dtype=ClassType.NETWORK)
backbone_cls.freeze()
if getattr(backbone_cls, 'out_channels') and 'in_channels' not in neck:
neck_in_channel = backbone_cls.out_channels
params = {"in_channels": neck_in_channel}
neck_cls = self.define_props('neck',
neck,
dtype=ClassType.NETWORK,
params=params)
else:
neck_cls = self.define_props('neck', neck, dtype=ClassType.NETWORK)
backbone_neck = Sequential(backbone_cls, neck_cls)
backbone_neck.freeze()
self.model = ClassFactory.get_cls(ClassType.NETWORK,
network_name)(backbone_neck,
num_classes, **kwargs)
def get_module_class(cls_name):
"""Get Search Space by class name.
:param cls_name: class name
:return: Search Space cls
"""
return ClassFactory.get_cls(ClassType.SEARCH_SPACE, cls_name)
def __new__(cls, *args, **kwargs):
"""Create optimizer or multi-optimizer class."""
if isinstance(cls.config.to_dict, list):
t_cls = ClassFactory.get_cls(ClassType.OPTIMIZER,
'MultiOptimizers')
return super().__new__(t_cls)
return super().__new__(cls)
def __init__(self, metric_cfg=None):
"""Init Metrics."""
self.mdict = {}
metric_config = self.config.to_dict() if not metric_cfg else deepcopy(metric_cfg)
if not isinstance(metric_config, list):
metric_config = [metric_config]
for metric_item in metric_config:
ClassFactory.get_cls(ClassType.METRIC, self.config.type)
metric_name = metric_item.pop('type')
metric_class = ClassFactory.get_cls(ClassType.METRIC, metric_name)
if isfunction(metric_class):
metric_class = partial(metric_class, **metric_item.get("params", {}))
else:
metric_class = metric_class(**metric_item.get("params", {}))
self.mdict[metric_name] = metric_class
self.mdict = Config(self.mdict)
def __init__(self, aux_weight, loss_base):
"""Init MixAuxiliaryLoss."""
super(MixAuxiliaryLoss, self).__init__()
self.aux_weight = aux_weight
loss_base_cp = loss_base.copy()
loss_base_name = loss_base_cp.pop('type')
self.loss_fn = ClassFactory.get_cls('trainer.loss', loss_base_name)(**loss_base_cp['params'])
def _init_transforms(self):
"""Initialize transforms method.
:return: a list of object
:rtype: list
"""
if "transforms" in self.args.keys():
transforms = list()
if not isinstance(self.args.transforms, list):
self.args.transforms = [self.args.transforms]
for i in range(len(self.args.transforms)):
transform_name = self.args.transforms[i].pop("type")
kwargs = self.args.transforms[i]
if ClassFactory.is_exists(ClassType.TRANSFORM, transform_name):
transforms.append(
ClassFactory.get_cls(ClassType.TRANSFORM,
transform_name)(**kwargs))
else:
transforms.append(
getattr(
importlib.import_module('torchvision.transforms'),
transform_name)(**kwargs))
return transforms
else:
return list()
def decorator(cls):
"""Provide input param to decorator.
:param func: wrapper function
:return: decoratpr
"""
# TODO： 需要导入包
if isinstance(class_name, str):
need_validate_cls = ClassFactory.get_cls(ClassType.CONFIG,
class_name)
else:
need_validate_cls = class_name
@wraps(cls)
def wrapper(*args, **kwargs):
"""Make function as a wrapper."""
valid_attrs = {
key: item
for key, item in cls.__dict__.items()
if not key.startswith('_')
}
for attr_name, rules in valid_attrs.items():
attr_value = getattr(need_validate_cls, attr_name)
if isinstance(rules, list) or isinstance(rules, tuple):
for _rule in rules:
_rule(attr_value)
else:
rules(attr_value)
return cls(*args, **kwargs)
return wrapper
def _create_loss(self):
"""Create loss class."""
if self.loss is None:
return
loss_cls = ClassFactory.get_cls(ClassType.LOSS, self.loss)
desc = self.desc.get('loss')
loss_obj = loss_cls(**desc) if desc is not None else loss_cls()
self.add_loss(loss_obj)
def _init_dataloader(self, mode, loader=None):
"""Init dataloader."""
if loader is not None:
return loader
dataset_cls = ClassFactory.get_cls(ClassType.DATASET)
dataset = dataset_cls(mode=mode)
dataloader = Adapter(dataset).loader
return dataloader
def __init__(self):
super(LatencyFilter, self).__init__()
self.max_latency = self.restrict_config.latency
if self.max_latency is not None:
dataset_cls = ClassFactory.get_cls(ClassType.DATASET)
self.dataset = dataset_cls()
from zeus.datasets import Adapter
self.dataloader = Adapter(self.dataset).loader
def _dispatch_trainer(self, samples):
for (id, desc, hps) in samples:
cls_trainer = ClassFactory.get_cls(ClassType.TRAINER)
TrainerConfig.from_dict(self.user_trainer_config)
trainer = cls_trainer(id=id, model_desc=desc, hps=hps)
evaluator = self._get_evaluator(trainer)
logging.info("submit trainer, id={}".format(id))
ReportServer.add_watched_var(General.step_name, trainer.worker_id)
self.master.run(trainer, evaluator)
def _add_pretrained_hook(self):
if self.pretrained_hook is None:
return
if isinstance(self.pretrained_hook, str):
hook = ClassFactory.get_cls(ClassType.PRETRAINED_HOOK,
self.pretrained_hook)
else:
hook = self.pretrained_hook
self._register_load_state_dict_pre_hook(hook)
def __init__(self):
"""Initialize."""
# register pytorch/tensorflow optim as default
raw_config = self.config.to_json()
raw_config.type = self.config.type
map_dict = OptimMappingDict
self.map_config = ConfigBackendMapping(
map_dict.type_mapping_dict, map_dict.params_mapping_dict).backend_mapping(raw_config)
self.optim_cls = ClassFactory.get_cls(ClassType.OPTIMIZER, self.map_config.type)
def __init__(self):
"""Initialize."""
# register pytorch loss as default
raw_config = self.config.to_json()
raw_config.type = self.config.type
map_dict = LossMappingDict()
self.map_config = ConfigBackendMapping(
map_dict.type_mapping_dict, map_dict.params_mapping_dict).backend_mapping(raw_config)
self._cls = ClassFactory.get_cls(ClassType.LOSS, self.map_config.type)
def _get_evaluator(self, worker_id):
if not PipeStepConfig.evaluator_enable:
return None
cls_evaluator = ClassFactory.get_cls('evaluator', "Evaluator")
evaluator = cls_evaluator({
"step_name": self.task.step_name,
"worker_id": worker_id
})
return evaluator
def value(self):
"""Get values."""
value = self._values.get(self.key)
if self.d_type == ClassType.NETWORK:
if isinstance(value, str):
cls = ClassFactory.get_cls(ClassType.NETWORK, value)
value = cls() if self.params is None else cls(**self.params)
else:
value = ClassFactory.get_instance(ClassType.NETWORK, value)
return value
def do(self):
"""Do the main task in this pipe step."""
logger.info("SpNasPipeStep started")
while not self.generator.is_completed:
id, spnas_sample = self.generator.search_alg.search()
cls_trainer = ClassFactory.get_cls('trainer')
trainer = cls_trainer(spnas_sample=spnas_sample, id=id)
logging.info("submit trainer(id={})!".format(id))
self.master.run(trainer)
self.master.join()
def _use_evaluator(self):
"""Check if use evaluator and get the evaluators.
:return: if we used evaluator, and Evaluator classes
:rtype: bool, (Evaluator, HostEvaluator, DloopEvaluator)
"""
use_evaluator = False
cls_evaluator_set = []
if EvaluatorConfig.host_evaluator_enable:
cls_host_evaluator = ClassFactory.get_cls(ClassType.HOST_EVALUATOR, "HostEvaluator")
use_evaluator = True
cls_evaluator_set.append(cls_host_evaluator)
if EvaluatorConfig.device_evaluator_enable:
cls_device_evaluator = ClassFactory.get_cls(
ClassType.DEVICE_EVALUATOR, "DeviceEvaluator")
use_evaluator = True
cls_evaluator_set.append(cls_device_evaluator)
# TODO HAVA_D_EVALUATOR
return use_evaluator, cls_evaluator_set
def create_module(model):
"""Create search space from model or desc."""
if isinstance(model, Module):
return model.__class__.__name__, model
elif isinstance(model, dict):
module_type = model.get('type')
module_param = deepcopy(model)
module_param.pop('type')
module = ClassFactory.get_cls(ClassType.SEARCH_SPACE, module_type)
return module_type, module(**module_param)
def _dispatch_trainer(self, samples):
for (id_ele, desc) in samples:
hps = deepcopy(desc)
cls_trainer = ClassFactory.get_cls(ClassType.TRAINER)
trainer = cls_trainer(id=id_ele, model_desc=desc, hps=hps)
evaluator = self._get_evaluator(trainer)
logging.info("submit trainer, id={}".format(id_ele))
self.master.run(trainer, evaluator)
if isinstance(samples, list) and len(samples) > 1:
self.master.join()
def _use_evaluator(self):
"""Check if use evaluator and get the evaluators.
:return: if we used evaluator, and Evaluator classes
:rtype: bool, (Evaluator, GpuEvaluator, DloopEvaluator)
"""
use_evaluator = False
cls_evaluator_set = []
if EvaluatorConfig.gpu_evaluator_enable:
cls_gpu_evaluator = ClassFactory.get_cls(ClassType.GPU_EVALUATOR,
"GpuEvaluator")
use_evaluator = True
cls_evaluator_set.append(cls_gpu_evaluator)
if EvaluatorConfig.davinci_mobile_evaluator_enable:
cls_davinci_mobile_evaluator = ClassFactory.get_cls(
ClassType.DAVINCI_MOBILE_EVALUATOR, "DavinciMobileEvaluator")
use_evaluator = True
cls_evaluator_set.append(cls_davinci_mobile_evaluator)
# TODO HAVA_D_EVALUATOR
return use_evaluator, cls_evaluator_set