vm.caffe.layers

Classes

class Accuracy : Compute the top-k accuracy.

class ArgMax : Compute the indices of maximum elements along the given axis.

class BatchNorm : Apply the batch normalization. [Ioffe & Szegedy, 2015].

class Cast : Cast the data type of input.

class Concat : Concatenate the inputs along the given axis.

class Convolution : Apply the n-dimension convolution.

class Crop : Select the elements according to the dimensions of second bottom.

class Data : Load batch of data for image classification.

class Deconvolution : Apply the n-dimension deconvolution.

class DepthwiseConv2d : Apply the 2d depthwise convolution. [Chollet, 2016].

class Dropout : Set the elements of the input to zero randomly. [Srivastava et.al, 2014].

class Eltwise : Compute the element-wise operation on the sequence of inputs.

class ELU : Apply the exponential linear unit. [Clevert et.al, 2015].

class ExpandDims : Expand a new dimension with size 1 at the given axis.

class EuclideanLoss : Compute the element-wise squared error.

class Flatten : Flatten the input along the given axes.

class FusedBatchNorm : Apply the fused batch normalization. [Ioffe & Szegedy, 2015].

class FusedGroupNorm : Apply the fused group normalization. [Wu & He, 2018].

class GroupNorm : Apply the group normalization. [Wu & He, 2018].

class InnerProduct : Compute the dense matrix multiplication along the given axes.

class Input : Produce input blobs with shape and dtype.

class LRN : Apply the local response normalization. [Krizhevsky et.al, 2012].

class Normalize : Apply the fused L2 normalization. [Liu et.al, 2015].

class Permute : Permute the dimensions of input.

class Pooling : Apply the n-dimension pooling.

class Power : Compute the power of input.

class PReLU : Apply the parametric rectified linear unit. [He et.al, 2015].

class Python : Wrap a python class into a layer.

class Reduction : Compute the reduction value along the given axis.

class ReLU : Apply the rectified linear unit. [Nair & Hinton, 2010].

class Reshape : Change the dimensions of input.

class ROIAlign : Apply the average roi align. [He et.al, 2017].

class ROIPooling : Apply the max roi pooling. [Girshick, 2015].

class Scale : Compute the affine transformation along the given axes.

class SELU : Apply the scaled exponential linear unit. [Klambauer et.al, 2017].

class Sigmoid : Apply the sigmoid function.

class SigmoidCrossEntropyLoss : Compute the sigmoid cross entropy with contiguous targets.

class SmoothL1Loss : Compute the element-wise error transited from L1 and L2. [Girshick, 2015].

class Softmax : Apply the softmax function.

class SoftmaxWithLoss : Compute the softmax cross entropy with sparse labels.

class TanH : Apply the tanh function.

class Tile : Tile the input according to the given multiples.