tiny_dnn  1.0.0
A header only, dependency-free deep learning framework in C++11
Class List
Here are the classes, structs, unions and interfaces with brief descriptions:
[detail level 1234]
 Nmodels
 Calexnet
 Ntiny_dnn
 Nactivation
 Ncore
 Ndetail
 Nweight_init
 CDevice
 CProgram
 CProgramHash
 CProgramManager
 CTensor
 CConv2dGradOp
 CConv2dOp
 CConv2dLibDNNForwardOp
 CConv2dLibDNNBackwardOp
 CConv2dOpenCLForwardOp
 CConv2dOpenCLBackwardOp
 CFullyConnectedGradOp
 CFullyConnectedOp
 CMaxPoolGradOp
 CMaxPoolOp
 Ctimer
 Cprogress_display
 Celementwise_add_layerElement-wise add N vectors y_i = x0_i + x1_i + ..
 Caverage_pooling_layerAverage pooling with trainable weights
 Caverage_unpooling_layerAverage pooling with trainable weights
 Cbatch_normalization_layerBatch Normalization
 Cconcat_layerConcat N layers along depth
 Cconvolutional_layer2D convolution layer
 Cdeconvolutional_layer2D deconvolution layer
 Cdropout_layerApplies dropout to the input
 Cfeedforward_layerSingle-input, single-output network with activation function
 Cfully_connected_layerCompute fully-connected(matmul) operation
 Cinput_layer
 ClayerBase class of all kind of NN layers
 Clinear_layerElement-wise operation: f(x) = h(scale*x+bias)
 Clrn_layerLocal response normalization
 Cmax_pooling_layerApplies max-pooing operaton to the spatial data
 Cmax_unpooling_layerApplies max-pooing operaton to the spatial data
 Cpartial_connected_layer
 Cpower_layerElement-wise pow: y = scale*x^factor
 Cquantized_convolutional_layer2D convolution layer
 Cquantized_deconvolutional_layer2D deconvolution layer
 Cquantized_fully_connected_layerCompute fully-connected(matmul) operation
 Cslice_layerSlice an input data into multiple outputs along a given slice dimension
 Cmse
 Cabsolute
 Cabsolute_eps
 Ccross_entropy
 Ccross_entropy_multiclass
 Cresult
 CnetworkA model of neural networks in tiny-dnn
 CnodeBase class of all kind of tinny-cnn data
 CedgeClass containing input/output data
 Cnode_tuple
 CnodesBasic class of various network types (sequential, multi-in/multi-out)
 CsequentialSingle-input, single-output feedforward network
 CgraphGeneric graph network
 CoptimizerBase class of optimizer usesHessian : true if an optimizer uses hessian (2nd order derivative of loss function)
 Cstateful_optimizer
 CadagradAdaptive gradient method
 CRMSpropRMSprop
 Cadam[a new optimizer (2015)]
 Cgradient_descentSGD without momentum
 CmomentumSGD with momentum
 Caligned_allocator
 Cdeserialization_helper
 Cgraph_visualizerUtility for graph visualization
 CimageSimple image utility class
 Cnn_errorError exception class for tiny-dnn
 Cnn_warnWarning class for tiny-dnn (for debug)
 Cnn_infoInfo class for tiny-dnn (for debug)
 Cnn_not_implemented_error
 Cblocked_range
 Crandom_generator
 Cserialization_helper
 Cindex3d
 Nvectorize
 Ndetail
 Cfoobar