Class Hierarchy

This inheritance list is sorted roughly, but not completely, alphabetically:

[detail level 123]

Backend
keraflow.backend.tensorflow_backend.TensorflowBackend
keraflow.backend.theano_backend.TheanoBackend
Exception
keraflow.utils.exceptions.KeraFlowError
MultiInputLayer
keraflow.layers.core.Concatenate	Concatenate multiple input tensors
keraflow.layers.core.ElementWiseMult	Reduce multiple input tensors by conducting multiplication operation
keraflow.layers.core.ElementWiseSum	Reduce multiple input tensors by conducting summation operation
object
keraflow.backend.common.Backend
keraflow.callbacks.Callback	Abstract base class used to build new callbacks
keraflow.callbacks.EarlyStopping	Stop training when a monitored quantity has stopped improving
keraflow.callbacks.LearningRateScheduler	Learning rate scheduler
keraflow.callbacks.ModelCheckpoint	Save the model after every epoch
keraflow.constraints.MaxNorm	Constrain the weights along an axis pattern to have unit norm
keraflow.constraints.NonNeg	Constrain the weights to be non-negative
keraflow.constraints.UnitNorm	Constrain the weights along an axis pattern to have unit norm
keraflow.layers.base.Kensor	Wrapper for a backend tensor
keraflow.layers.base.Input	The entering point of each model
keraflow.layers.base.Layer	Building block of a model
keraflow.layers.base.Input	The entering point of each model
keraflow.layers.base.MultiInputLayer	Base class for layer accepting multiple inputs
keraflow.layers.base.SequentialLayer	Class for making Sequential a Layer
keraflow.layers.wrappers.TimeDistributed	Wrapper for apply a layer to every temporal slice of an input
keraflow.models.Model	A model is a directed Kensor graph
keraflow.models.Sequential	Model with single input and single output
keraflow.optimizers.Optimizer	Abstract optimizer base class
keraflow.optimizers.Adadelta	Adadelta optimizer
keraflow.optimizers.Adagrad	Adagrad optimizer
keraflow.optimizers.Adam	Adam optimizer
keraflow.optimizers.Adamax	Adamax optimizer from Adam paper's Section 7
keraflow.optimizers.RMSprop	RMSProp optimizer
keraflow.optimizers.SGD	Stochastic gradient descent, with support for momentum, learning rate decay, and Nesterov momentum
keraflow.regularizers.L1	L1 weight regularization penalty, also known as LASSO
keraflow.regularizers.L1L2	L1-L2 weight regularization penalty, also known as ElasticNet
keraflow.regularizers.L2	L2 weight regularization penalty, also known as weight decay, or Ridge
keraflow.utils.user_input_utils.UserInput	Utility class for flexible user input assiging optimizers, regularizers, numpy input..
SequentialLayer
keraflow.models.Sequential	Model with single input and single output
Layer
keraflow.layers.convolution.Convolution3D	Not implemented yet
keraflow.layers.convolution.ConvolutionBase	Base layer for convolution layers
keraflow.layers.convolution.Convolution1D	Convolution layer for convolving (sequence_length, input_dim) inputs
keraflow.layers.convolution.Convolution2D	Convolution layer for convolving (input_depth, input_row, input_col) inputs
keraflow.layers.convolution.PoolingBase	Base layer for pooling layers
keraflow.layers.convolution.Pooling1D	Pooling layer for sub-sampling (sequence_length, input_dim) inputs
keraflow.layers.convolution.Pooling2D	Pooling layer for sub-sampling (input_depth, input_row, input_col) inputs
keraflow.layers.convolution.Pooling3D	Zero-padding layer for (input_depth, input_x, input_y, input_z) inputs
keraflow.layers.convolution.UnSamplingBase	Base layer for unsampling layers
keraflow.layers.convolution.UnSampling1D	Repeat each temporal step length times along the time axis
keraflow.layers.convolution.UnSampling2D	Unsampling layer for (input_depth, input_row, input_col) inputs
keraflow.layers.convolution.UnSampling3D	Unsampling layer for (input_depth, input_x, input_y, input_z) inputs
keraflow.layers.convolution.ZeroPaddingBase	Base layer for zero padding layers
keraflow.layers.convolution.ZeroPadding1D	Zero-padding layer for (sequence_length, input_dim) inputs
keraflow.layers.convolution.ZeroPadding2D	Zero-padding layer for (input_depth, input_row, input_col) inputs
keraflow.layers.convolution.ZeroPadding3D	Zero-padding layer for (input_depth, input_x, input_y, input_z) inputs
keraflow.layers.core.Activation	Applies an activation function to an output
keraflow.layers.core.Dense	Fully connected layer
keraflow.layers.core.Dropout	Applies Dropout to the input
keraflow.layers.core.ExpandDims	Expand dimension of the input tensor
keraflow.layers.core.Flatten	Flatten the input tensor into 1D
keraflow.layers.core.Highway	Densely connected highway network
keraflow.layers.core.Lambda	Wrapper for implementating simple inline layer
keraflow.layers.core.PermuteDims	Permutes the dimensions of the input tensor according to a given pattern
keraflow.layers.core.Repeat	Repeat the input tensor n times along given axis
keraflow.layers.core.Reshape	Reshapes the input tensor according to a given pattern
keraflow.layers.embeddings.Embedding	Vocabulary (row) vectors looking up layer
keraflow.layers.recurrent.Recurrent	Base class for recurrent layers
keraflow.layers.recurrent.GRU	Gated Recurrent Unit - Cho et al
keraflow.layers.recurrent.LSTM	Long-Short Term Memory unit - Hochreiter 1997
keraflow.layers.recurrent.SimpleRNN	Fully-connected RNN where the output is to be fed back to input