caffe2.python.model_helper.ModelHelper Class Reference

Inheritance diagram for caffe2.python.model_helper.ModelHelper:

Public Member Functions
def	__init__ (self, name=None, init_params=True, allow_not_known_ops=True, skip_sparse_optim=False, param_model=None, arg_scope=None)
def	arg_scope (self)
def	get_name (self)
def	create_param (self, param_name, shape, initializer, tags=None)
def	get_param_info (self, param)
def	add_param_DEPRECATED (self, param, key=None, shape=None, length=None)
def	param_info (self, grad_type=None, id=None)
def	AddParameter (self, param, tags=None)
def	GetParams (self, namescope=None, top_scope=False)
def	Proto (self)
def	InitProto (self)
def	RunAllOnGPU (self, args, kwargs)
def	CreateDB (self, blob_out, db, db_type, kwargs)
def	AddGradientOperators (self, args, kwargs)
def	get_param_to_grad (self, params)
def	GetOptimizationParamInfo (self, params=None)
def	Validate (self)
def	GetComputedParams (self, namescope=None)
def	GetAllParams (self, namescope=None)
def	TensorProtosDBInput (self, unused_blob_in, blob_out, batch_size, db, db_type, kwargs)
def	GetDevices (self)
def	__getattr__ (self, op_type)
def	__dir__ (self)

Public Attributes
	name
	net
	param_init_net
	param_to_grad
	params
	gradient_ops_added
	init_params
	allow_not_known_ops
	skip_sparse_optim
	weights
	biases
	grad_map

Detailed Description

A helper model so we can manange models more easily. It contains net def
and parameter storages. You can add an Operator yourself, e.g.

    model = model_helper.ModelHelper(name="train_net")
    # init your weight and bias as w and b
    w = model.param_init_net.XavierFill(...)
    b = model.param_init_net.ConstantFill(...)
    fc1 = model.FC([input, w, b], output, **kwargs)

or you can use helper functions in brew module without manually
defining parameter initializations and operators.

    model = model_helper.ModelHelper(name="train_net")
    fc1 = brew.fc(model, input, output, dim_in, dim_out, **kwargs)

Definition at line 74 of file model_helper.py.

Member Function Documentation

def caffe2.python.model_helper.ModelHelper.__getattr__	(		self,
			op_type
	)

Catch-all for all other operators, mostly those without params.

Definition at line 444 of file model_helper.py.

def caffe2.python.model_helper.ModelHelper.create_param	(		self,
			param_name,
			shape,
			initializer,
			tags = None
	)

Creates parameter with a given name and initializer.

If param_name is instance of BlobRefernce - then this blob will be used
to store parameter (no any logic will affect it's location).

If param_name is instance of a string type, then the final blob will
be created in the CurrentNameScope with the respect of all parameter
sharing logic, i.e. 'resolved_name_scope/param_name'.

Parameter sharing logic is going to override CurrentNameScope accoring
to the rules that are specified through ParameterSharing contexts,
all ParameterSharing contexts are applied recursively until there are no
extra overrides present, where on each step the best match will be
applied first.

The following examples should clarify the way ParameterSharing logic
works:

As an example if this function is called with parameter 'w':
a. Call from some scope 'global_scope' with no Parameter sharing:
  'global_scope/w'
b. Call from scope 'scope_b', with override {'scope_b': 'scope_a'}:
  'scope_a/w'
c. Call from scope 'scope_a', with override {'scope_a': ''}:
  'scope_a/w'
d. Call from scope 'scope_b/shared', with overrides
  {'scope_b/shared': 'scope_b', 'scope_b': 'scope_a'}:
  'scope_a/w'
d. Call from scope 'scope_b/unshared', with overrides
  {'scope_b/shared': 'scope_b', 'scope_b': 'scope_a'}:
  'scope_a/unshared/w'

Definition at line 160 of file model_helper.py.

def caffe2.python.model_helper.ModelHelper.get_param_to_grad	(		self,
			params
	)

Given a list of parameters returns a dict from a parameter
to a corresponding gradient

Definition at line 353 of file model_helper.py.

def caffe2.python.model_helper.ModelHelper.GetComputedParams	(		self,
			namescope = None
	)

Returns the computed params in current namescope. 'Computed params'
are such parameters that are not optimized via gradient descent but are
directly computed from data, such as the running mean and variance
of Spatial Batch Normalization.

Definition at line 410 of file model_helper.py.

def caffe2.python.model_helper.ModelHelper.GetOptimizationParamInfo	(		self,
			params = None
	)

Returns a map for param => grad.
If params is not specified, all parameters will be considered.

Definition at line 369 of file model_helper.py.

def caffe2.python.model_helper.ModelHelper.GetParams	(		self,
			namescope = None,
			top_scope = False
	)

Returns the params in current namescope

Definition at line 297 of file model_helper.py.

def caffe2.python.model_helper.ModelHelper.TensorProtosDBInput	(		self,
			unused_blob_in,
			blob_out,
			batch_size,
			db,
			db_type,
			kwargs
	)

TensorProtosDBInput.

Definition at line 430 of file model_helper.py.

---

The documentation for this class was generated from the following file:

• caffe2/python/model_helper.py