|View source on GitHub|
Flattens the input while maintaining the batch_size.
tf.contrib.layers.flatten( inputs, outputs_collections=None, scope=None )
Assumes that the first dimension represents the batch.
||A tensor of size [batch_size, ...].|
||Collection to add the outputs.|
||Optional scope for name_scope.|
|A flattened tensor with shape [batch_size, k].|
||If inputs rank is unknown or less than 2.|