def GraphDef_Update(meta_graph, updater):
"""Inject the update targets into GraphDef.
The ``updater`` should generate update targets before.
Parameters
----------
meta_graph : dragon_pb2.GraphDef
The definition of meta graph.
updater : BaseUpdater
The updater.
Returns
-------
None
"""
if updater is None: return
updater._prefix = meta_graph.name + '_'
extra_arguments = updater._extra_kwargs
extra_arguments['domain'] = updater._prefix
parallel_arguments = {}
# wrap hyper-parameters as Tensor for CC
for k, v in updater._hyper_params.items():
ws.FeedTensor(updater._prefix + k, np.array([v], dtype=np.float32))
# check data parallel if necessary
if mpi.Is_Init():
idx, group = mpi.AllowParallel()
if idx != -1:
parallel_arguments['parallel_mode'] = mpi.GetParallelMode()
parallel_arguments['comm'], parallel_arguments['group'] \
= mpi.CreateGroup(root=group[0], incl=group)
parallel_arguments['root'] = group[0]
for k, v in parallel_arguments.items():
meta_graph.arg.add().CopyFrom(MakeArgument(k, v))
for tuple in updater._tuples:
tensors = tuple[0]
arguments = tuple[1]
kwargs = dict(arguments, **extra_arguments)
u_target = pb.UpdateTarget()
u_target.type = updater._type
_, u_target.name = GetOperatorName()
for tensor in tensors:
u_target.tensor.append(tensor)
for k, v in kwargs.items():
u_target.arg.add().CopyFrom(MakeArgument(k, v))
meta_graph.u_target.extend([u_target])
def GraphDef_Update(meta_graph, updater):
"""Inject the update targets into GraphDef.
The ``updater`` should generate update targets before.
Parameters
----------
meta_graph : dragon_pb2.GraphDef
The definition of meta graph.
updater : BaseUpdater
The updater.
Returns
-------
None
"""
if updater is None: return
# use graph name if missing slot
if updater._slot is None:
updater._slot = meta_graph.name
extra_arguments = updater._extra_kwargs
extra_arguments['slot'] = updater._slot
parallel_arguments = {}
updater.register_in_workspace()
# check data parallel if necessary
if mpi.Is_Init():
idx, group = mpi.AllowParallel()
if idx != -1:
parallel_arguments['parallel_mode'] = mpi.GetParallelMode()
parallel_arguments['comm'], parallel_arguments['group'] \
= mpi.CreateGroup(root=group[0], incl=group)
parallel_arguments['root'] = group[0]
for k, v in parallel_arguments.items():
meta_graph.arg.add().CopyFrom(MakeArgument(k, v))
for e in updater._param_group:
pair, arguments = e
kwargs = dict(arguments, **extra_arguments)
u_target = pb.UpdateTarget()
u_target.type = updater.type()
_, u_target.name = GetOperatorName()
for t in pair:
u_target.tensor.append(t)
for k, v in kwargs.items():
u_target.arg.add().CopyFrom(MakeArgument(k, v))
meta_graph.u_target.extend([u_target])
def GraphDef_Phase(meta_graph, targets):
"""Inject the phase into GraphDef.
If existing gradients, we assume it should be ``TRAIN``, and vice versa.
Parameters
----------
meta_graph : dragon_pb2.GraphDef
The definition of meta graph.
targets : list
The solving targets.
Returns
-------
None
"""
phase = 'TEST'
from dragon.core.scope import _PHASE_SCOPE
if _PHASE_SCOPE != '':
phase = _PHASE_SCOPE.upper()
else:
for target in targets:
if len(target.grad_wrts) > 0:
phase = 'TRAIN'
break
meta_graph.arg.extend([MakeArgument('phase', phase)])
def GraphDef_Update(graph_def, updater):
""" generate all update targets for CC Graph """
if updater is None: return
updater._prefix = graph_def.name + '_'
extra_kwargs = updater._extra_kwargs
extra_kwargs['domain'] = updater._prefix
# wrap hyper-parameters as Tensor for CC
for k, v in updater._hyper_params.items():
ws.FeedTensor(updater._prefix + k, np.array([v], dtype=np.float32))
# check data parallel if necessary
if mpi.is_init():
idx, group = mpi.allow_parallel()
if idx != -1:
extra_kwargs['comm'], extra_kwargs['group'] \
= mpi.group(root=group[0], incl=group)
extra_kwargs['root'] = group[0]
extra_kwargs['mode'] = mpi.get_parallel_mode()
extra_kwargs['group_size'] = len(group)
for tuple in updater._tuples:
tensors = tuple[0]
kwargs = tuple[1]
kwargs = dict(kwargs, **extra_kwargs)
u_target = pb.UpdateTarget()
u_target.type = updater._type
_, u_target.name = GetOperatorName()
for tensor in tensors:
u_target.tensor.append(tensor)
for k, v in kwargs.items():
u_target.arg.add().CopyFrom(MakeArgument(k, v))
graph_def.u_target.extend([u_target])
def GraphDef_Opt(meta_graph):
"""Inject the optimization options into GraphDef.
Parameters
----------
meta_graph : dragon_pb2.GraphDef
The definition of meta graph.
Returns
-------
None
References
----------
`config.SetDebugMode(*args, **kwargs)`_ - How the enable debug mode.
`memonger.share_grads(*args, **kwargs)`_ - How the enable gradients sharing.
"""
from dragon.config import option
OX = 3 if option['share_grads'] else 2
if option['debug_mode']: OX = 1
meta_graph.arg.add().CopyFrom(MakeArgument('optimization_level', OX))
meta_graph.graph_type = option['graph_type']
def GraphDef_Phase(graph_def, targets):
phase = 'TEST'
from dragon.core.scope import PHASE_SCOPE
global PHASE_SCOPE
if PHASE_SCOPE != '': phase = PHASE_SCOPE.upper()
else:
for target in targets:
if len(target.grad_wrts) > 0:
phase = 'TRAIN'
break
graph_def.arg.extend([MakeArgument('phase', phase)])