MyCaffe  1.11.7.7
Deep learning software for Windows C# programmers.
MyCaffe.layers.beta Namespace Reference

The MyCaffe.layers.ssd namespace contains all SSD related layers. More...

Classes

class  AccuracyDecodeLayer
 The AccuracyDecodeLayer compares the labels output by the DecodeLayer with the expected labels output by the DataLayer. This layer is initialized with the MyCaffe.param.AccuracyParameter. More...
 
class  AccuracyEncodingLayer
 The AccuracyEncodingLayer computes the classification accuracy for an encoding used in a classification task that uses a Siamese Network or similar type of net that creates an encoding mapped to a label. This layer is initialized with the MyCaffe.param.AccuracyParameter. More...
 
class  ConvolutionOctaveLayer
 The ConvolutionOctaveLayer processes high and low frequency portions of images using convolution. More...
 
class  DataSequenceLayer
 DataSequence Layer - this caches inputs by label and then outputs data item tuplets that include an 'anchor', optionally a 'positive' match, and at least one 'negative' match. More...
 
class  DecodeLayer
 The DecodeLayer decodes the label of a classification for an encoding produced by a Siamese Network or similar type of net that creates an encoding mapped to a set of distances where the smallest distance indicates the label for which the encoding belongs. More...
 
class  GatherLayer
 The GatherLayer extracts (gathers) data from specified indices along a given axis from the input and returns it as the output. The indices are passed in as the second bottom blob. More...
 
class  GRNLayer
 The GRNLayer performs an L2 normalization over the input data. More...
 
class  InterpLayer
 The InterpLayer changes the spatial resolution by bi-linear interpolation. More...
 
class  KnnLayer
 
class  LayerFactory
 The LayerFactor is responsible for creating all layers implemented in the MyCaffe.layers.ssd namespace. More...
 
class  MeanErrorLossLayer
 The MeanErrorLossLayer computes losses based on various different Mean Error methods as shown below. This layer is used to solve regression problems such as time-series predictions. More...
 
class  MergeLayer
 The MergeLayer merges two bottom blobs with a specified copy pattern and outputs a single blob result. More...
 
class  MishLayer
 The MishLayer provides a novel activation function that tends to work better than ReLU. This layer is initialized with the MyCaffe.param.MishParameter. More...
 
class  ModelDataLayer
 The ModelDataLayer loads data from RawImageResults table for an encoder/decoder type model. More...
 
class  Normalization1Layer
 The Normalization1Layer performs an L2 normalization over the input data. This layer is initialized with the MyCaffe.param.Normalization1Parameter. More...
 
class  SerfLayer
 The SerfLayer provides a novel activation function that tends to work better than ReLU. More...
 
class  SqueezeLayer
 The SqueezeLayer performs a squeeze operation where all single dimensions are removed. More...
 
class  TextDataLayer
 The TextDataLayer loads data from text data files for an encoder/decoder type model. This layer is initialized with the MyCaffe.param.TextDataParameter. More...
 
class  TransposeLayer
 The TransposeLayer performs a permute and transpose operation similar to numpy.transpose. More...
 
class  TripletLossLayer
 TripletLoss Layer - this is the triplet loss layer used to calculate the triplet loss and gradients using the triplet loss method of learning. The triplet loss method involves image triplets using the following format: Anchor (A), Positives (P) and Negatives (N) More...
 
class  UnPoolingLayer
 
class  UnPoolingLayer1
 
class  UnsqueezeLayer
 The UnsqueezeLayer performs an unsqueeze operation where a single dimension is inserted at each index specified. More...
 

Detailed Description

The MyCaffe.layers.ssd namespace contains all SSD related layers.