API reference¶
Subpackages¶
Base classes¶
-
class
deeppy.base.
Model
[source]¶ Bases:
deeppy.base.ParamMixin
Fillers¶
-
class
deeppy.filler.
AutoFiller
(gain=1.0)[source]¶ Bases:
deeppy.filler.Filler
-
class
deeppy.filler.
ConstantFiller
(value=0.0)[source]¶ Bases:
deeppy.filler.Filler
-
class
deeppy.filler.
CopyFiller
(np_array)[source]¶ Bases:
deeppy.filler.Filler
-
class
deeppy.filler.
NormalFiller
(mu=0.0, sigma=1.0)[source]¶ Bases:
deeppy.filler.Filler
-
class
deeppy.filler.
UniformFiller
(low, high)[source]¶ Bases:
deeppy.filler.Filler
Parameters¶
-
class
deeppy.parameter.
Parameter
(fill, name='', learn_rate=1.0, weight_decay=0.0, monitor=False)[source]¶ Bases:
deeppy.base.PickleMixin
-
array
¶
-
grad
()[source]¶ Returns a parameter step calculated from the gradient. This differs from grad_array() as the parameter may be shared such that its gradient has multiple sources.
-
grad_array
¶ Returns the gradient array.
-
Bases:
deeppy.parameter.Parameter