RKNN2 V2.5.0 · RV1103B/RV1106B：Recurrent networks, arithmetic and pooling

This RKNN2 V2.5.0 platform reference helps ArmSoM developers select and adapt models. It does not establish support on every ArmSoM board: verify the board SoC, actual Toolkit2, Runtime, NPU driver and BSP versions. The source specifies no universal Linux kernel version. No ArmSoM hardware validation was performed.

Source: the supplied Chinese and English RKNN2 Supported Operators and Constraints Reference V2.5.0, 374 and 384 pages respectively. Chinese follows the Chinese edition; English, Russian and German follow the English edition. Platform constraints remain separate. The English PDF itself contains Chinese I/O figures and output tables; their provenance is retained with editable localized tables. Source spellings such as ConvTranpose and the original LSTM/GRU equations are preserved; these equations are not corrections to SDK implementations.

Source pages: CN 324–331; EN 334–341.

Rockchip RKNN2

Operator reference series

RKNN2 V2.5.0 · RK3566/RK3568：Basics and convolution

RKNN2 V2.5.0 · RK3566/RK3568：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RK3566/RK3568：Shapes, slicing and transpose

RKNN2 V2.5.0 · RK3566/RK3568：Activations, normalization and extensions

RKNN2 V2.5.0 · RK3588：Basics and convolution

RKNN2 V2.5.0 · RK3588：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RK3588：Shapes, slicing and transpose

RKNN2 V2.5.0 · RK3588：Activations, normalization and extensions

RKNN2 V2.5.0 · RK3562：Basics and convolution

RKNN2 V2.5.0 · RK3562：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RK3562：Shapes, slicing and transpose

RKNN2 V2.5.0 · RK3562：Activations, normalization and extensions

RKNN2 V2.5.0 · RV1103/RV1106：Basics and convolution

RKNN2 V2.5.0 · RV1103/RV1106：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RV1103/RV1106：Shapes, slicing and transpose

RKNN2 V2.5.0 · RV1103/RV1106：Activations, normalization and extensions

RKNN2 V2.5.0 · RK3576/RV1126B：Basics and convolution

RKNN2 V2.5.0 · RK3576/RV1126B：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RK3576/RV1126B：Shapes, slicing and transpose

RKNN2 V2.5.0 · RK3576/RV1126B：Activations, normalization and extensions

RKNN2 V2.5.0 · RK2118：Basics and convolution

RKNN2 V2.5.0 · RK2118：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RK2118：Shapes, slicing and transpose

RKNN2 V2.5.0 · RK2118：Activations, normalization and extensions

RKNN2 V2.5.0 · RV1103B/RV1106B：Basics and convolution

RKNN2 V2.5.0 · RV1103B/RV1106B：Recurrent networks, arithmetic and pooling

RKNN2 V2.5.0 · RV1103B/RV1106B：Shapes, slicing and transpose

RKNN2 V2.5.0 · RV1103B/RV1106B：Activations, normalization and extensions

RKNN2 V2.5.0 · CPU operator support and constraints

RKNN2 V2.5.0 · GPU operator support and constraints

RKNN2 V2.5.0 · Model I/O, alignment and convolution limits

Div

Performs element-wise binary division (C=A/B), supporting multi-directional NumPy-style broadcasting

Inputs

• A:T1,T2
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
  • Broadcasting support
    • [b,c,1,1]
    • [scalar]
• B:T1,T2
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
  • Broadcasting support
    • [b,c,1,1]
    • [scalar]

Outputs

• C:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer

Data Type Constraints

• T1:float16(Tensor)
• T2:float(constTensor)

Expand

Broadcast the input tensor according to the given shape and broadcasting rules. The broadcasting rule is similar to numpy.array(input) * numpy.ones(shape): dimensions are right- aligned; two corresponding dimensions must have the same value, or one of them equals 1

Inputs

• input:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
• shape:T2
  • shape:[batch,channel,height,width]

Outputs

• output:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer

Data Type Constraints

• T1:int8(Tensor)
• T2:int64(Tensor)

Broadcasting constraints

• [b,c,1,w]->[b,c,h,w],h∈[1,8192]
• [b,1,1,w]->[b,c,h,w],h∈[1,8192]
• [b,c,h,1]->[b,c,h,w],w ∈[1,8192]
• [b,1,h,1]->[b,c,h,w],w ∈[1,8192]
• [b,c,1,1]->[b,c,h,w],h∈[1,8192],w ∈[1,8192]

Max

Computes the element-wise maximum value of the input tensor

Inputs

• A:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
  • Broadcasting support
    • Supports broadcasting along arbitrary dimensions
    • Broadcasting constraints are the same as the Expand operator
• B:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
  • Broadcasting support
    • Supports broadcasting along arbitrary dimensions
    • Broadcasting constraints are the same as the Expand operator

Outputs

• C:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer/per-channel

Data Type Constraints

• T1:int8(Tensor)

Other Support

• Multi-core: Not yet supported

Mul

Performs element-wise binary multiplication (C=A*B), supporting multi-directional NumPy-style broadcasting

Inputs

• A:T1,T2
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
  • Broadcasting support
    • Supports broadcasting along arbitrary dimensions
    • Broadcasting constraints are the same as the Expand operator
• B:T1,T2
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
  • Broadcasting support
    • Supports broadcasting along arbitrary dimensions
    • Broadcasting constraints are the same as the Expand operator

Outputs

• C:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer

Data Type Constraints

• T1:int8(Tensor)
• T2:float(constTensor)

MulRelu

Mul and Relu fused computation, C=Relu(A*B)

Inputs

• Same as Mul

Outputs

• Same as Mul

Other Support

• Multi-core: Supported

Pad

Given a tensor (data) containing the data to be padded, a tensor (pads) containing the start and end padding values for each axis, a mode (mode), and a constant value (constant_value), generate a padded tensor (output)

Inputs

• data:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer
• pads:T2
  • shape : [batch_begin, channel_begin, height_begin, width_begin, batch_end, channel_end,height_end,width_end]
• constant_value(optional):T3
  • shape:[1]
  • Quantization support
    • per-layer

Outputs

• output:T1
  • shape:[batch,channel,height,width]
  • Quantization support
    • per-layer

Data Type Constraints

• T1:int8(Tensor)
• T2:int64(Tensor)
• T3:float(Scalar),int8(Scalar),float16(Scalar)

Attributes

• mode: string (default value constant)
  • Supported modes: constant, reflect
    • constantNo limitation
    • reflectchannel∈(0,8192],height∈(0,8192],width∈(0,8176]

Other Support

• Multi-core: Not yet supported

MaxPool

MaxPool consumes an input tensor X and applies max pooling to the tensor according to kernel size, stride size, and padding length. Max pooling consists of computing the maximum over all values of a subset of the input tensor according to the kernel size, and downsampling the data into output tensor Y for further processing. The output spatial shape calculation differs depending on whether explicit padding (when using pads) or automatic padding (when using auto_pad) is used. Only explicit padding is used

Inputs

• input:T1
  • shape:[batch,channel,height,width]
    • channel∈(0,8192]
    • height∈(0,7]
    • width∈(0,7]

Outputs

• output:T1
  • shape:[batch,channel,height,width]

Attributes

• kernel_shape:int64(kernel_h,kernel_w)
  • kernel_h∈(0,7]
  • kernel_w∈(0,7]
• auto_pad:string
  • Padding mode: only NOTSET is supported
• ceil_mode:int64
  • Whether to use ceil or floor to compute the output shape: not supported
• strides:int64(strides_h,strides_w)
  • strides_h∈(0,8]
  • strides_w∈(0,8]
• count_include_pad:int64[]
  • count_include_pad, whether to include pad values in computation: 1
• pad:int64(pad_top,pad_left,pad_bottom,pad_right)
  • pad_top∈[0,7]
  • pad_left∈[0,7]
  • pad_bottom∈[0,7]
  • pad_right∈[0,7]

Data Type Constraints

• T1:int8(Tensor)
• T2:float(Tensor)
• T3:float(Scalar)

GlobalMaxPool

GlobalMaxPool consumes an input tensor X and applies max pooling over the values in the same channel. This is equivalent to a MaxPool with kernel_shape equal to the spatial dimensions of the input.

Inputs

• input:T1
  • shape:[batch,channel,height,width]
    • channel∈(0,8192]
    • height∈(0,7]
    • width∈(0,7]

Outputs

• output:T1
  • shape:[batch,channel,height,width]

Attributes

• kernel_shape:int64(kernel_h,kernel_w)
  • kernel_h∈(0,7]
  • kernel_w∈(0,7]
• auto_pad:string
  • Padding mode: only NOTSET is supported
• ceil_mode:int64
  • Whether to use ceil or floor to compute the output shape: not supported
• strides:int64(strides_h,strides_w)
  • strides_h∈(0,8]
  • strides_w∈(0,8]
• count_include_pad:int64[]
  • count_include_pad, whether to include pad values in computation: 1
• pad:int64(pad_top,pad_left,pad_bottom,pad_right)
  • pad_top∈[0,7]
  • pad_left∈[0,7]
  • pad_bottom∈[0,7]
  • pad_right∈[0,7]

Data Type Constraints

• T1:int8(Tensor)
• T2:float(Tensor)
• T3:float(Scalar)

AveragePool

• Not supported (converted to convolution implementation on the tool side)

GlobalAveragePool

• Not supported (converted to convolution implementation on the tool side)