def flatten(inputs, name=None):
"""
Flattens the input while maintaining the batch_size.
Assumes that the first dimension represents the batch.
Args:
inputs: A tensor of size [batch_size, ...].
scope: Optional name for operation.
Returns:
A flattened tensor with shape [batch_size, k].
"""
return ops.Flatten(inputs, axis=1)
def flatten(x, outdim=1):
"""Flatten the input by keeping the specific dimensions.
Parameters
----------
x : Tensor
The input tensor.
outdim : int
The number of dimensions to keep.
Returns
-------
Tensor
The output tensor.
"""
return ops.Flatten(x, keep_axes=outdim)
def to_one_hot(y, nb_class, **kwargs):
"""Generate a matrix where each row corresponds to the one hot encoding.
The ``y`` should be a 1d vector.
Parameters
----------
y: Tensor
The input tensor.
nb_class : int
The number of classes.
Returns
-------
Tensor
The one hot matrix.
"""
flat_y = ops.Flatten(y, keep_axes=1)
return ops.OneHot(flat_y, depth=nb_class)
def flatten(inputs, name=None):
return ops.Flatten(inputs, axis=1)
def LayerSetup(self, bottom):
return _ops.Flatten(bottom, **self.arguments)
def flatten(inputs, outputs_collections=None, scope=None):
return op_lib.Flatten(inputs, axis=0, keep_axes=2)
def Setup(self, bottom):
super(FlattenLayer, self).Setup(bottom)
input = bottom[0] if isinstance(bottom, list) else bottom
return ops.Flatten(input, **self._param)
