SMAUG
Simulating Machine Learning Applications on gem5-Aladdin
Classes
All SMAUG Operators

All operators supported by SMAUG. More...

Classes

class  smaug::BatchNormOp< Backend >
 Implements the batch normalization layer. More...
 
class  smaug::ConcatOp< Backend >
 Concatenates N Tensors along a specified axis. More...
 
class  smaug::SwitchOp< Backend >
 Conditionally forwards an input to one of two outputs. More...
 
class  smaug::MergeOp< Backend >
 Forwards the first live input to its output. More...
 
class  smaug::ConvolutionOp< Backend >
 The base class for all 4D spatial convolution operators. More...
 
class  smaug::DataOp< Backend >
 Exposes a Tensor as its only output. More...
 
class  smaug::DepthwiseConvolutionOp< Backend >
 Implements the depthwise convolution operator. More...
 
class  smaug::EltwiseAddOp< Backend >
 Adds two Tensors elementwise. More...
 
class  smaug::EltwiseMulOp< Backend >
 Multiplies two Tensors elementwise. More...
 
class  smaug::EltwiseOp< Backend >
 The base class of all elementwise operators. More...
 
class  smaug::EluOp< Backend >
 Implements the exponential linear unit function. More...
 
class  smaug::SeluOp< Backend >
 Implements the scaled exponential linear unit function. More...
 
class  smaug::FusedActivationOp
 An Operator fused with an activation function. More...
 
class  smaug::GreaterOp< Backend >
 Implements an elementwise greater than operator. More...
 
class  smaug::GreaterEqualOp< Backend >
 Implements an elementwise greater than or equal to operator. More...
 
class  smaug::InnerProductOp< Backend >
 Implements the inner product operator. More...
 
class  smaug::LessOp< Backend >
 Implements an elementwise less-than operator. More...
 
class  smaug::LessEqualOp< Backend >
 Implements an elementwise less-than-or-equal-to operator. More...
 
class  smaug::PaddingOp< Backend >
 Pad a given tensor in any number of dimensions with arbitrary size. More...
 
class  smaug::PoolingOp< Backend >
 Implements a pooling operator. More...
 
class  smaug::MaxPoolingOp< Backend >
 Implements the max-pooling operator. More...
 
class  smaug::AvgPoolingOp< Backend >
 Implements the arithmetic-average-pooling operator. More...
 
class  smaug::ReluOp< Backend >
 Implements the rectified linear unit operator: max(slope * x, 0). More...
 
class  smaug::ReorderOp< Backend >
 Implements a Tensor reordering operation to convert between different DataLayouts. More...
 
class  smaug::RepeatOp< Backend >
 Replicates a Tensor's data among all dimensions. More...
 
class  smaug::ReshapeOp< Backend >
 Changes the Tensor's shape while retaining the number of elements. More...
 
class  smaug::SigmoidOp< Backend >
 Implements the sigmoid operator, defined as 1/(1 + exp(-input)). More...
 
class  smaug::SoftmaxOp< Backend >
 Implements the softmax operator. More...
 
class  smaug::SplitOp< Backend >
 Implements the split operator, which divides a Tensor into N output Tensors along a specified dimension. More...
 
class  smaug::TanhOp< Backend >
 Implements the tanh operator. More...
 
class  smaug::HardTanhOp< Backend >
 Implements the hard tanh operator, which bounds the min and max value of the tanh operator. More...
 
class  smaug::UnaryOp< Backend >
 Base class for all operators with one input. More...
 

Detailed Description

All operators supported by SMAUG.