728x90
반응형

 

 

https://docs.nvidia.com/deeplearning/sdk/tensorrt-support-matrix/index.html#layers-precision-matrix

 

TensorRT Support Matrix :: Deep Learning SDK Documentation

These support matrices provide a look into the supported platforms, features, and hardware capabilities of the TensorRT 6.0.1 APIs, parsers, and layers. For previously released TensorRT documentation, see TensorRT Archives.

docs.nvidia.com

 

 

 

3.1. TensorFlow Container 19.07-19.08 (TensorFlow 1.14)
Acos
Acosh
ArgMax
ArgMin
Asin
Asinh
Atan
Ceil
ClipByValue
CombinedNonMaxSuppression
Conv2DBackpropInput
Cos
Cosh
DepthToSpace
Elu
Floor
FusedBatchNormV3
GatherV2
LeakyRelu
Pack
Selu
Sin
Sinh
Softplus
Softsign
SpaceToDepth
Split
SquaredDifference
StopGradient
Tanh
Transpose
Unpack

 


3.2. TensorFlow Container 19.02-19.06 (TensorFlow 1.13)
ExpandDims
FakeQuantWithMinMaxArgs
FakeQuantWithMinMaxVars
Pow
QuantizeAndDequantizeV2
QuantizeAndDequantizeV3
Reshape
Sigmoid
Slice
Sqrt
Square
Squeeze
StridedSlice
Tanh

 


3.3. TensorFlow Container 18.11-19.01 (TensorFlow 1.12)
Abs
Add
AvgPool
BatchMatMul
BiasAdd
ConcatV2
Const
Conv2D
DepthwiseConv2dNative
Div
Exp
FusedBatchNorm
FusedBatchNormV2
Identity
Log
MatMul
Max
Maximum
MaxPool
Mean
Min
Minimum
Mul
Neg
Pad
Prod
RealDiv
Reciprocal
Relu
Relu6
Rsqrt
Snapshot
Softmax
Sub
Sum
TopKV2

728x90
반응형