You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
classModel(object):
defget_model_output(self, inputs: theano.Varable) ->theano.Variable
@abstractpropertydefparameters(self) ->Iterable[theano.SharedVariable]
@abstractmethoddefsave(self, path)
@abstractclassmethoddefload(self, path)
@abstractmethoddefinitialize(self, rng)
# A random number generator (instead of a seed) to generate the model with default initializersdefuse_regression(self, X: theano.Variable) ->theano.Variablereturnself._use_regression(self.generate_model_output(X))
defuse_classification(self, X: theano.Variable) ->theano.Variablereturnself._use_classification(self.generate_model_output(X))
@abstractmethoddef_use_regression(self, out: theano.Variable) ->theano.Variable
@abstractmethoddef_use_classification(self, out: theano.Variable) ->theano.Variable
We should discuss the interface of
Model
, what method should be mandatory for subclasses.Definitions (to be added in a wiki)
Model
:Some questions
Initialize
in the model. If so, what should it took as input?Suggestions
The text was updated successfully, but these errors were encountered: