# Migrating from Hexagon-nn

Developers familiar with Hexagon-nn APIs can follow this short guide to learn how to translate
Hexagon-nn API function calls into the equivalent QNN API function calls.

Both Hexagon-nn and QNN follow a similar process to create and execute a neural network (graph).

![../../_static/resources/qnn_tutorial_migrate_process.png](data:image/png;base64,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)

The examples below will use the following format to show comparisons of **Hexagon-nn** vs **QNN**
code blocks. Notable APIs will be highlighted within the code blocks.

Hexagon-nn Initialize

1hexagon_nn_init((hexagon_nn_nn_id *) &nng_id);
    Copy to clipboard

QNN Backend Creation

1QnnBackend_create(Qnn_LogHandle_t logger, const (QnnBackend_Config_t **) config, (Qnn_BackendHandle_t*) backend);
    Copy to clipboard

In the examples that follow, error handling is shown but not in all cases to limit the amount of
code clutter in the examples. A full solution would require more robust error handling which is
left as an exercise for the developer.

To illustrate the various APIs the following simple graph (migration) will serve as an example for
this tutorial.

![../../_static/resources/qnn_tutorial_migrate_tiny.png](data:image/png;base64,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)

Note

This tutorial assumes that you are familiar with all the concepts required to create
and run a quantized neural network with Hexagon-nn. The topic of quanitized networks,
input and output tensors will not be covered here. For the purposes of this guide
all tensor data is assumed to be in a quantized format.

## Create Graph

Create an instance of the neural network graph.

Hexagon-nn Create Graph

1 hexagon_nn_nn_id nng_id = 0;
     2
     3 if (hexagon_nn_config() != 0)
     4 {
     5     printf("hexagon_nn_config failed.\n");
     6 }
     7 else if (hexagon_nn_init((hexagon_nn_nn_id *) &nng_id))
     8 {
     9     printf("hexagon_nn_init failed.\n");
    10 }
    Copy to clipboard

Graphs for QNN require the following objects to be created:

> 
> 
> - log (optional)
> - backend
> - device
> - context
> - graph (associated with context)

QNN Create Graph

1Qnn_LogHandle_t logger{nullptr};
     2Qnn_BackendHandle_t backend{nullptr};
     3Qnn_DeviceHandle_t device{nullptr};
     4
     5QnnLog_Callback_t logCallBack{nullptr};
     6if (QNN_SUCCESS != QnnLog_create(logCallBack, QNN_LOG_LEVEL_ERROR, &logger)) {
     7   printf("QNN log creation failed.\n");
     8}
     9const QnnBackend_Config_t *backendConfigs{nullptr};
    10if (QNN_SUCCESS != QnnBackend_create(logger, &backendConfigs, &backend)) {
    11   printf("QNN backend creation failed.\n");
    12} else {
    13   const QnnDevice_Config_t *deviveConfigs{nullptr};
    14   if (QNN_SUCCESS != QnnDevice_create(logger, &deviveConfigs, &device)) {
    15      printf("QNN device creation failed.\n");
    16   } else {
    17      Qnn_ContextHandle_t context{nullptr};
    18      const QnnContext_Config_t *contextConfigs{nullptr};
    19      if (QNN_SUCCESS != QnnContext_create(backend, device, &contextConfigs, &context)) {
    20         printf("QNN context creation failed.\n");
    21      } else {
    22         // choose any name to identify the graph
    23         const char* graphName = "migrate";
    24         const QnnGraph_Config_t *graphConfigs{nullptr};
    25         Qnn_GraphHandle_t graph{nullptr};
    26         if (QNN_SUCCESS != QnnGraph_create(context, graphName, &graphConfigs, &graph)) {
    27            printf("QNN graph creation failed.\n");
    28         }
    29      }
    30   }
    31}
    Copy to clipboard

## Add Nodes

This next section will cover adding nodes to the graph. Nodes are added with the following APIs.

Hexagon-nn Add Nodes

1int hexagon_nn_append_node( ... );
    2int hexagon_nn_append_const_node( ... );
    Copy to clipboard

QNN Add Nodes

1Qnn_ErrorHandle_t QnnTensor_createGraphTensor( ... );
    2Qnn_ErrorHandle_t QnnGraph_addNode( ... );
    Copy to clipboard

Nodes in Hexagon-nn graphs use unique IDs to identify graph nodes. Node weights, biases, and
parameters are generally added as constant nodes. The diagram below shows the simple example graph
with Hexagon-nn unique IDs.

![../../_static/resources/qnn_tutorial_migrate_hexnn.png](data:image/png;base64,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)

In QNN nodes the weights, biases and parameters are added as tensors with static data.
The diagram below shows the simple example graph with QNN names to help identify each node.

![../../_static/resources/qnn_tutorial_migrate_qnn.png](data:image/png;base64,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)

### Adding Tensors and Nodes

Each node operation is associated with a set of input and output tensors. The node operation
definition defines the type and ordering of the inputs and outputs.

**Hexagon-nn**

A node can be a constant node or a graph node. Nodes are added to the graph via
hexagon\_nn\_append\_node or hexagon\_nn\_append\_const\_node.

For example,

Hexagon-nn Add Constant Node

1 // add constant nodes
    2 static uint8_t weights[108] = { 219, 39, 82, 157, 172, 80, ...};
    3
    4 ret = hexagon_nn_append_const_node(nng_id, 0x12001, 3, 3, 3, 4, (const uint8_t *) weights, 108);
    Copy to clipboard

Each graph node is supplied with a set of inputs and outputs. These will include the constant
nodes and activation tensors. Each of these are identified by a node ID and an index value.

These inputs and outputs are specified as lists when adding nodes into the graph. The lists
contain node ID references which are used to create the connectivity of the graph.

Hexagon-nn Add Graph Node

1 // define the input
     2 hexagon_nn_input inputs[] = {
     3                               { // output of the INPUT node
     4                                   .src_id = 0x01000,
     5                                   .output_idx = 0,
     6                               },
     7                               { // weights
     8                                   .src_id = 0x12001,
     9                                   .output_idx = 0,
    10                               },
    11
    12                               ...
    13                             };
    14
    15 // define the output
    16 hexagon_nn_output outputs[] = {
    17                                 {.rank = 4,
    18                                  .max_sizes = {1, 30, 30, 4, 0, 0, 0, 0},
    19                                  .elementsize = 4,
    20                                  .zero_offset = 0,
    21                                  .stepsize = 0.0},
    22
    23                                 ...
    24
    25                                 {.rank = 4,
    26                                  .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    27                                  .elementsize = 4,
    28                                  .zero_offset = 0,
    29                                  .stepsize = 0.0}
    30                               };
    31
    32 // add the node to the graph
    33 ret = hexagon_nn_append_node(nng_id, 0x02000, OP_QuantizedConv2d_8x8to32, NN_PAD_VALID, inputs, 7, outputs, 3);
    Copy to clipboard

**QNN**

The Qnn\_Tensor\_t data structure is used to define the attributes when creating a tensor.
Before a node can be created, the set of activation and parameter tensors connected to it
need to be created.

Note that the tensor ID does not need to be set. When calling the QnnTensor\_createGraphTensor
function, a tenor ID will be generated and written back to tensor.v2.id. And creating a tensor
with a name that duplicates a previously created tensor name in the graph results in undefined
behaviour.

For example,

QNN Create Tensors

1 // input (activation tensor)
     2 uint32_t inputDimensions[] = {1, 32, 32, 3};
     3 Qnn_Tensor_t inputTensor                                 = QNN_TENSOR_INIT;
     4 inputTensor.version                                      = QNN_TENSOR_VERSION_2;
     5 inputTensor.v2.id                                        = 0;
     6 inputTensor.v2.name                                      = "input";
     7 inputTensor.v2.type                                      = QNN_TENSOR_TYPE_APP_WRITE;
     8 inputTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     9 inputTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    10 inputTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    11 inputTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    12 inputTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0256975870579481;
    13 inputTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -132;
    14 inputTensor.v2.rank                                      = 4;
    15 inputTensor.v2.dimensions                                = inputDimensions;
    16 inputTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    17 inputTensor.v2.clientBuf                                 = QNN_CLIENT_BUFFER_INIT
    18 inputTensor.v2.isDynamicDimensions                       = nullptr;
    19 inputTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    20 inputTensor.v2.isProduced                                = 0;
    21
    22 ret = QnnTensor_createGraphTensor(graph, &inputTensor);
    23
    24 // weights (static tensor)
    25 static uint8_t weights[] = {232, 177, 126, 160, 15, 171, ...};
    26 uint32_t weightsDimensions[] = {3, 3, 3, 4};
    27 Qnn_Tensor_t weightsTensor                                 = QNN_TENSOR_INIT;
    28 weightsTensor.version                                      = QNN_TENSOR_VERSION_2;
    29 weightsTensor.v2.id                                        = 0;
    30 weightsTensor.v2.name                                      = "weights";
    31 weightsTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
    32 weightsTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    33 weightsTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    34 weightsTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    35 weightsTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    36 weightsTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0012544898781925;
    37 weightsTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -129;
    38 weightsTensor.v2.rank                                      = 4;
    39 weightsTensor.v2.dimensions                                = weightsDimensions;
    40 weightsTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    41 weightsTensor.v2.clientBuf.data                            = weights;
    42 weightsTensor.v2.clientBuf.dataSize                        = 108;
    43 weightsTensor.v2.isDynamicDimensions                       = nullptr;
    44 weightsTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    45 weightsTensor.v2.isProduced                                = 0;
    46
    47 ret = QnnTensor_createGraphTensor(graph, &weightsTensor);
    48
    49 // output (activation tensor)
    50 uint32_t outputDimensions[] = {1, 30, 30, 4};
    51 Qnn_Tensor_t outputTensor                                 = QNN_TENSOR_INIT;
    52 outputTensor.version                                      = QNN_TENSOR_VERSION_2;
    53 outputTensor.v2.id                                        = 0;
    54 outputTensor.v2.name                                      = "output";
    55 outputTensor.v2.type                                      = QNN_TENSOR_TYPE_APP_READ;
    56 outputTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    57 outputTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    58 outputTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    59 outputTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    60 outputTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0166513100266457;
    61 outputTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -107;
    62 outputTensor.v2.rank                                      = 4;
    63 outputTensor.v2.dimensions                                = outputDimensions;
    64 outputTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    65 outputTensor.v2.clientBuf                                 = QNN_CLIENT_BUFFER_INIT;
    66 outputTensor.v2.isDynamicDimensions                       = nullptr;
    67 outputTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    68 outputTensor.v2.isProduced                                = 0;
    69
    70 ret = QnnTensor_createGraphTensor(graph, &outputTensor);
    Copy to clipboard

Unlike Hexagon-nn where the weights and weights min/max are set in multiple constant nodes, QNN
directly includes the quantization scale and offset values during tensor creation.

Parameters for a node are defined via an array Qnn\_Param\_t parameters, each of which are uniquely
named. The parameter values can apply to tensors or scalars, with specific values defined in the
operation definitions.

QNN Params Tensor

1 // parameter tensors
     2 uint32_t pad[] = {0, 0, 0, 0};
     3 uint32_t padDimensions[] = {2, 2};
     4 Qnn_Tensor_t padTensor                                 = QNN_TENSOR_INIT;
     5 padTensor.version                                      = QNN_TENSOR_VERSION_2;
     6 padTensor.v2.id                                        = 0;
     7 padTensor.v2.name                                      = "pad";
     8 padTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
     9 padTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    10 padTensor.v2.dataType                                  = QNN_DATATYPE_UINT_32;
    11 padTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_UNDEFINED;
    12 padTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_UNDEFINED;
    13 padTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0;
    14 padTensor.v2.quantizeParams.scaleOffsetEncoding.offset = 0;
    15 padTensor.v2.rank                                      = 2;
    16 padTensor.v2.dimensions                                = padDimensions;
    17 padTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    18 padTensor.v2.clientBuf.data                            = pad;
    19 padTensor.v2.clientBuf.dataSize                        = 16;
    20 padTensor.v2.isDynamicDimensions                       = nullptr;
    21 padTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    22 padTensor.v2.isProduced                                = 0;
    23
    24 ret = QnnTensor_createGraphTensor(graph, &padTensor);
    25
    26 Qnn_Param_t nodeParams[1] = {QNN_PARAM_INIT};
    27 nodeParams[0].paramType   = QNN_PARAMTYPE_TENSOR;
    28 nodeParams[0].name        = QNN_OP_CONV_2D_PARAM_PAD_AMOUNT;
    29 nodeParams[0].tensorParam = padTensor;
    Copy to clipboard

After the set up of the parameters, static and activation tensors,
the next step is to add the node to the graph.

The Qnn\_OpConfig\_t data structure is used to define the graph node when it is created.

QNN Add Node

1 // create a list of all input tensors
     2 Qnn_Tensor_t inputList[2] = {QNN_TENSOR_INIT, QNN_TENSOR_INIT};
     3 inputList[0]              = inputTensor;
     4 inputList[1]              = weightsTensor;
     5
     6 // create a list of all output tensors
     7 Qnn_Tensor_t outputList[1] = {QNN_TENSOR_INIT};
     8 outputList[0]              = outputTensor;
     9
    10 // operation attributes
    11 Qnn_OpConfig_t opDefinition   = QNN_OPCONFIG_INIT;
    12 opDefinition.v1.name          = "exampleOp";
    13 opDefinition.v1.packageName   = "qti.aisw";
    14 opDefinition.v1.typeName      = QNN_OP_XYZ;
    15 opDefinition.v1.numOfParams   = 1;
    16 opDefinition.v1.params        = nodeParams;
    17 opDefinition.v1.numOfInputs   = 2;
    18 opDefinition.v1.inputTensors  = inputList;
    19 opDefinition.v1.numOfOutputs  = 1;
    20 opDefinition.v1.outputTensors = outputList;
    21
    22 // add the node to the graph
    23 ret = QnnGraph_addNode(graph, opDefinition);
    Copy to clipboard

Now with a comparative understanding of how nodes are added to Hexagon-nn and QNN the next section
will illustrate adding nodes to the simple network to show the connectivity.

### Add Nodes - input

**Hexagon-nn Input**

![../../_static/resources/qnn_tutorial_migrate_input_hnn.png](data:image/png;base64,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)

Hexagon-nn requires an INPUT op to be added with the output describing the dimensions and sizes of
the graph’s input(s).  It will read in tensors from hexagon\_nn\_tensordef during the
hexagon\_nn\_execute call. See the [Execute Graph](https://docs.qualcomm.com/doc/80-63442-50/topic/hexnn_migration.html#execute-graph) section below for more details.

Hexagon-nn Input

1 hexagon_nn_output outputs[] = {{.rank = 4,
     2                                 .max_sizes = {1, 32, 32, 3, 0, 0, 0, 0},
     3                                 .elementsize = 1,
     4                                 .zero_offset = 0,
     5                                 .stepsize = 0.0},
     6                                {.rank = 4,
     7                                 .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
     8                                 .elementsize = 4,
     9                                 .zero_offset = 0,
    10                                 .stepsize = 0.0},
    11                                {.rank = (4),
    12                                 .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    13                                 .elementsize = 4,
    14                                 .zero_offset = 0,
    15                                 .stepsize = 0.0}};
    16
    17 hexagon_nn_append_node(nng_id, 0x01000, OP_INPUT, NN_PAD_NA, 0, 0, outputs, 3);
    Copy to clipboard

**QNN Input**

![../../_static/resources/qnn_tutorial_migrate_input_qnn.png](data:image/png;base64,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)

Note

The QNN input is an activation tensor defined to be of type QNN\_TENSOR\_TYPE\_APP\_WRITE.

QNN Input

1 // input (activation tensor)
     2 uint32_t inputDimensions[] = {1, 32, 32, 3}
     3 Qnn_Tensor_t inputTensor                                 = QNN_TENSOR_INIT;
     4 inputTensor.version                                      = QNN_TENSOR_VERSION_2;
     5 inputTensor.v2.id                                        = 0;
     6 inputTensor.v2.name                                      = "convInput";
     7 inputTensor.v2.type                                      = QNN_TENSOR_TYPE_APP_WRITE;
     8 inputTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     9 inputTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    10 inputTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    11 inputTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    12 inputTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0256975870579481;
    13 inputTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -132;
    14 inputTensor.v2.rank                                      = 4;
    15 inputTensor.v2.dimensions                                = inputDimensions;
    16 inputTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    17 inputTensor.v2.clientBuf                                 = QNN_CLIENT_BUFFER_INIT;
    18 inputTensor.v2.isDynamicDimensions                       = nullptr;
    19 inputTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    20 inputTensor.v2.isProduced                                = 0;
    21
    22 ret = QnnTensor_createGraphTensor(graph, &inputTensor);
    Copy to clipboard

### Add Nodes - convolution

**Hexagon-nn Convolution**

![../../_static/resources/qnn_tutorial_migrate_conv_hnn.png](data:image/png;base64,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)

Hexagon-nn convolution node accepts 7 inputs

> 
> 
> - [3] input activation, min and max from the INPUT node
> - [3] weights, min and max added as constant nodes
> - [1] stride shape added as a constant node

and 3 outputs

> 
> 
> - [3] output activation, min and max

Hexagon-nn Convolution

1 // create weight constant nodes
     2 static uint8_t weights[108] = { 219, 39, 82, 157, 172, 80, ...};
     3 static float weightsMin[1] = { -0.003076 };
     4 static float weightsMax[1] = {  0.003076 };
     5
     6 ret = hexagon_nn_append_const_node(nng_id, 0x12001, 3, 3, 3, 4, (const uint8_t *) weights, 108);
     7 ret = hexagon_nn_append_const_node(nng_id, 0x12002, 1, 1, 1, 1, (const uint8_t *) weightsMin, 4);
     8 ret = hexagon_nn_append_const_node(nng_id, 0x12003, 1, 1, 1, 1, (const uint8_t *) weightsMax, 4);
     9
    10 // create stride constant node
    11 ret = hexagon_nn_append_const_node(nng_id, 0x12004, 1, 1, 1, 1, (const uint8_t *) 0, 0);
    12
    13 // define the input
    14 hexagon_nn_input convInputs[] = {
    15                               { // output of the INPUT node
    16                                   .src_id = 0x01000,
    17                                   .output_idx = 0,
    18                               },
    19                               { // weights
    20                                   .src_id = 0x12001,
    21                                   .output_idx = 0,
    22                               },
    23                               { // output of the INPUT min
    24                                   .src_id = 0x01000,
    25                                   .output_idx = 1,
    26                               },
    27                               { // output of the INPUT max
    28                                   .src_id = 0x01000,
    29                                   .output_idx = 2,
    30                               },
    31                               { // weights min
    32                                   .src_id = 0x12002,
    33                                   .output_idx = 0,
    34                               },
    35                               { // weights max
    36                                   .src_id = 0x12003,
    37                                   .output_idx = 0,
    38                               },
    39                               { // stride
    40                                   .src_id = 0x12004,
    41                                   .output_idx = 0,
    42                               }
    43                             };
    44
    45 // define the output
    46 hexagon_nn_output convOutputs[] = {
    47                                 {
    48                                   .rank = 4,
    49                                   .max_sizes = {1, 30, 30, 4, 0, 0, 0, 0},
    50                                   .elementsize = 4,
    51                                   .zero_offset = 0,
    52                                   .stepsize = 0.0
    53                                 },
    54                                 {
    55                                   .rank = 4,
    56                                   .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    57                                   .elementsize = 4,
    58                                   .zero_offset = 0,
    59                                   .stepsize = 0.0
    60                                 },
    61                                 {
    62                                   .rank = 4,
    63                                   .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    64                                   .elementsize = 4,
    65                                   .zero_offset = 0,
    66                                   .stepsize = 0.0
    67                                 }
    68                               };
    69
    70 // add the node to the graph
    71 ret = hexagon_nn_append_node(nng_id, 0x02000, OP_QuantizedConv2d_8x8to32, NN_PAD_VALID, convInputs, 7, convOutputs, 3);
    Copy to clipboard

Note

Padding is specified as an input argument into the hexagon\_nn\_append\_node API.

To add the biases after the convolution Hexagon-nn requires a biasAdd node to be appended to the
graph.

This bias node accepts 6 inputs

> 
> 
> - [3] input activation, min and max from output of the convolution node
> - [3] biases, min and max added as constant nodes

and 3 outputs

> 
> 
> - [3] output activation, min and max

![../../_static/resources/qnn_tutorial_migrate_conv_bias_hnn.png](data:image/png;base64,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)

Hexagon-nn Convolution biasAdd

1 // create bias constant nodes
     2 static int32_t convBias[10] = { -1315013842, ... , -2147483647 };
     3 static float convBiasMin[1] = { -0.860266387462616 };
     4 static float convBiasMax[1] = { 0.860266387462616 };
     5
     6 ret = hexagon_nn_append_const_node(nng_id, 0x13001, 3, 3, 3, 4, (const uint8_t *) convBias, 108);
     7 ret = hexagon_nn_append_const_node(nng_id, 0x13002, 1, 1, 1, 1, (const uint8_t *) convBiasMin, 4);
     8 ret = hexagon_nn_append_const_node(nng_id, 0x13003, 1, 1, 1, 1, (const uint8_t *) convBiasMax, 4);
     9
    10 hexagon_nn_input convBiasInputs[] = {
    11                                   {
    12                                     .src_id = 0x02000,
    13                                     .output_idx = 0,
    14                                   },
    15                                   {
    16                                      .src_id = 0x13001,
    17                                      .output_idx = 0,
    18                                   },
    19                                   {
    20                                      .src_id = 0x02000,
    21                                      .output_idx = 1,
    22                                   },
    23                                   {
    24                                      .src_id = 0x02000,
    25                                      .output_idx = 2,
    26                                   },
    27                                   {
    28                                      .src_id = 0x13002,
    29                                      .output_idx = 0,
    30                                   },
    31                                   {
    32                                      .src_id = 0x13003,
    33                                      .output_idx = 0,
    34                                   }
    35                                 };
    36
    37 hexagon_nn_output convBiasOutputs[] = {
    38                                     {
    39                                       .rank = 4,
    40                                       .max_sizes = {1, 1, 1, 10, 0, 0, 0, 0},
    41                                       .elementsize = 4,
    42                                       .zero_offset = 0,
    43                                       .stepsize = 0.0
    44                                     },
    45                                     {
    46                                       .rank = (4),
    47                                       .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    48                                       .elementsize = 4,
    49                                       .zero_offset = 0,
    50                                       .stepsize = 0.0
    51                                     },
    52                                     {
    53                                       .rank = (4),
    54                                       .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    55                                       .elementsize = 4,
    56                                       .zero_offset = 0,
    57                                       .stepsize = 0.0
    58                                     }
    59                                   };
    60
    61 ret = hexagon_nn_append_node(nng_id, 0x03000, OP_QuantizedBiasAdd_32p32to32,
    62                               NN_PAD_NA, convBiasInputs, 6, convBiasOutputs, 3)
    Copy to clipboard

For QNN, the biases are added as a static tensor input to the convolution node. *See convolutionexample below for details*.

**QNN Convolution**

![../../_static/resources/qnn_tutorial_migrate_conv_qnn.png](data:image/png;base64,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)

QNN convolution node accepts 4 inputs

> 
> 
> - [1] input activation tensor
> - [1] weights static tensor
> - [1] biases static tensor
> - [1] parameter list consisting of
> 
>     - [1] stride static tensor
>     - [1] pad static tensor

and 1 output:

> 
> 
> - [1] output activation

Note

See the [Finalize Graph](https://docs.qualcomm.com/doc/80-63442-50/topic/hexnn_migration.html#finalize-graph) section below for more details.

QNN Convolution

1 // output (activation tensor)
      2 uint32_t convOutputDimensions[] = {1, 30, 30, 4};
      3 Qnn_Tensor_t convOutputTensor                                 = QNN_TENSOR_INIT;
      4 convOutputTensor.version                                      = QNN_TENSOR_VERSION_2;
      5 convOutputTensor.v2.id                                        = 0;
      6 convOutputTensor.v2.name                                      = "convOutput";
      7 convOutputTensor.v2.type                                      = QNN_TENSOR_TYPE_NATIVE;
      8 convOutputTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
      9 convOutputTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
     10 convOutputTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
     11 convOutputTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
     12 convOutputTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0166513100266457;
     13 convOutputTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -107;
     14 convOutputTensor.v2.rank                                      = 4;
     15 convOutputTensor.v2.dimensions                                = convOutputDimensions;
     16 convOutputTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
     17 convOutputTensor.v2.clientBuf                                 = QNN_CLIENT_BUFFER_INIT;
     18 convOutputTensor.v2.isDynamicDimensions                       = nullptr;
     19 convOutputTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
     20 convOutputTensor.v2.isProduced                                = 0;
     21
     22 ret = QnnTensor_createGraphTensor(graph, &convOutputTensor);
     23
     24 // weights (static tensor)
     25 static uint8_t convWeights[] = {232, 177, 126, 160, 15, 171, ...};
     26 uint32_t convWeightsDimensions[] = {3, 3, 3, 4};
     27 Qnn_Tensor_t convWeightsTensor                                 = QNN_TENSOR_INIT;
     28 convWeightsTensor.version                                      = QNN_TENSOR_VERSION_2;
     29 convWeightsTensor.v2.id                                        = 0;
     30 convWeightsTensor.v2.name                                      = "convWeights";
     31 convWeightsTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
     32 convWeightsTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     33 convWeightsTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
     34 convWeightsTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
     35 convWeightsTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
     36 convWeightsTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0012544898781925;
     37 convWeightsTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -129;
     38 convWeightsTensor.v2.rank                                      = 4;
     39 convWeightsTensor.v2.dimensions                                = convWeightsDimensions;
     40 convWeightsTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
     41 convWeightsTensor.v2.clientBuf.data                            = convWeights;
     42 convWeightsTensor.v2.clientBuf.dataSize                        = 108;
     43 convWeightsTensor.v2.isDynamicDimensions                       = nullptr;
     44 convWeightsTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
     45 convWeightsTensor.v2.isProduced                                = 0;
     46
     47 ret = QnnTensor_createGraphTensor(graph, &convWeightsTensor);
     48
     49 // biases (static tensor)
     50 static uint8_t convBiases[] = {78, 255, 80, 157};
     51 uint32_t convBiasesDimensions[] = {4};
     52 Qnn_Tensor_t convBiasesTensor                                 = QNN_TENSOR_INIT;
     53 convBiasesTensor.version                                      = QNN_TENSOR_VERSION_2;
     54 convBiasesTensor.v2.id                                        = 0;
     55 convBiasesTensor.v2.name                                      = "convBiases";
     56 convBiasesTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
     57 convBiasesTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     58 convBiasesTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
     59 convBiasesTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
     60 convBiasesTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
     61 convBiasesTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0006514320266433;
     62 convBiasesTensor.v2.quantizeParams.scaleOffsetEncoding.offset = 0;
     63 convBiasesTensor.v2.rank                                      = 1;
     64 convBiasesTensor.v2.dimensions                                = convBiasesDimensions;
     65 convBiasesTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
     66 convBiasesTensor.v2.clientBuf.data                            = convBiases;
     67 convBiasesTensor.v2.clientBuf.dataSize                        = 4;
     68 convBiasesTensor.v2.isDynamicDimensions                       = nullptr;
     69 convBiasesTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
     70 convBiasesTensor.v2.isProduced                                = 0;
     71
     72 ret = QnnTensor_createGraphTensor(graph, &convBiasesTensor);
     73
     74 // stride (static tensor)
     75 uint32_t convStride[] = {1, 1};
     76 uint32_t convStrideDimensions[] = {2};
     77 Qnn_Tensor_t convStrideTensor                                 = QNN_TENSOR_INIT;
     78 convStrideTensor.version                                      = QNN_TENSOR_VERSION_2;
     79 convStrideTensor.v2.id                                        = 0;
     80 convStrideTensor.v2.name                                      = "convStride";
     81 convStrideTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
     82 convStrideTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     83 convStrideTensor.v2.dataType                                  = QNN_DATATYPE_UINT_32;
     84 convStrideTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_UNDEFINED;
     85 convStrideTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_UNDEFINED;
     86 convStrideTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0;
     87 convStrideTensor.v2.quantizeParams.scaleOffsetEncoding.offset = 0;
     88 convStrideTensor.v2.rank                                      = 1;
     89 convStrideTensor.v2.dimensions                                = convStrideDimensions;
     90 convStrideTensor.v2.clientBuf.data                            = convStride;
     91 convStrideTensor.v2.clientBuf.dataSize                        = 8;
     92 convStrideTensor.v2.isDynamicDimensions                       = nullptr;
     93 convStrideTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
     94 convStrideTensor.v2.isProduced                                = 0;
     95
     96 ret = QnnTensor_createGraphTensor(graph, &convStrideTensor);
     97
     98 // pad (static tensor)
     99 uint32_t convPad[] = {0, 0, 0, 0};
    100 uint32_t convPadDimensions[] = {2, 2};
    101 Qnn_Tensor_t convPadTensor                                 = QNN_TENSOR_INIT;
    102 convPadTensor.version                                      = QNN_TENSOR_VERSION_2;
    103 convPadTensor.v2.id                                        = 0;
    104 convPadTensor.v2.name                                      = "convPad";
    105 convPadTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
    106 convPadTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    107 convPadTensor.v2.dataType                                  = QNN_DATATYPE_UINT_32;
    108 convPadTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_UNDEFINED;
    109 convPadTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_UNDEFINED;
    110 convPadTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0;
    111 convPadTensor.v2.quantizeParams.scaleOffsetEncoding.offset = 0;
    112 convPadTensor.v2.rank                                      = 2;
    113 convPadTensor.v2.dimensions                                = convPadDimensions;
    114 convPadTensor.v2.clientBuf.data                            = convPad;
    115 convPadTensor.v2.clientBuf.dataSize                        = 16;
    116 convPadTensor.v2.isDynamicDimensions                       = nullptr;
    117 convPadTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    118 convPadTensor.v2.isProduced                                = 0;
    119
    120 ret = QnnTensor_createGraphTensor(graph, &convPadTensor);
    121
    122 Qnn_Param_t convParamsList[2] = {QNN_PARAM_INIT, QNN_PARAM_INIT};
    123 convParamsList[0].paramType   = QNN_PARAMTYPE_TENSOR;
    124 convParamsList[0].name        = QNN_OP_CONV_2D_PARAM_STRIDE;
    125 convParamsList[0].tensorParam = convStrideTensor;
    126 convParamsList[1].paramType   = QNN_PARAMTYPE_TENSOR;
    127 convParamsList[1].name        = QNN_OP_CONV_2D_PARAM_PAD_AMOUNT;
    128 convParamsList[1].tensorParam = convPadTensor;
    129
    130 // create a list of all input tensors
    131 Qnn_Tensor_t convInputList[3] = {
    132                                   inputTensor,
    133                                   convWeightsTensor,
    134                                   convBiasesTensor
    135                                 };
    136
    137 // create a list of all output tensors
    138 Qnn_Tensor_t convOutputList[1] = {convOutputTensor};
    139
    140 // operation definition
    141 Qnn_OpConfig_t convOpDefinition   = QNN_OPCONFIG_INIT;
    142 convOpDefinition.v1.name          = "conv";
    143 convOpDefinition.v1.packageName   = "qti.aisw";
    144 convOpDefinition.v1.typeName      = QNN_OP_CONV_2D;
    145 convOpDefinition.v1.numOfParams   = 2;
    146 convOpDefinition.v1.params        = convParamsList;
    147 convOpDefinition.v1.numOfInputs   = 3;
    148 convOpDefinition.v1.inputTensors  = convInputList;
    149 convOpDefinition.v1.numOfOutputs  = 1;
    150 convOpDefinition.v1.outputTensors = convOutputList;
    151
    152 // add the node to the graph
    153 ret = QnnGraph_addNode(graph, convOpDefinition);
    Copy to clipboard

### Add Nodes - fully connected (fc)

**Hexagon-nn FC**

![../../_static/resources/qnn_tutorial_migrate_fc_hnn.png](data:image/png;base64,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)

Hexagon-nn fully connected node is represented by the MatMul operation. It accepts 6 inputs.

> 
> 
> - [3] input activation, min and max from the output of the convolution biasAdd node
> - [3] weights, min and max added as constant nodes

and 3 outputs

> 
> 
> - [3] output activation, min and max

Hexagon-nn Fully Connected

1 // create bias constant nodes
     2 static int32_t bias[10] = { -1315013842, ... , -2147483647 };
     3 static float biasMin[1] = { -0.860266387462616 };
     4 static float biasMax[1] = { 0.860266387462616 };
     5
     6 ret = hexagon_nn_append_const_node(nng_id, 0x14001, 3, 3, 3, 4,
     7                                     (const uint8_t *) weights, 108);
     8
     9 ret = hexagon_nn_append_const_node(nng_id, 0x14002, 1, 1, 1, 1,
    10                                     (const uint8_t *) weightsMin, 4);
    11
    12 ret = hexagon_nn_append_const_node(nng_id, 0x14003, 1, 1, 1, 1,
    13                                     (const uint8_t *) weightsMax, 4);
    14
    15 hexagon_nn_input fcInputs[] = {
    16                                 {
    17                                   .src_id = 0x03000,
    18                                   .output_idx = 0,
    19                                 },
    20                                 {
    21                                    .src_id = 0x14001,
    22                                    .output_idx = 0,
    23                                 },
    24                                 {
    25                                    .src_id = 0x03001,
    26                                    .output_idx = 1,
    27                                 },
    28                                 {
    29                                    .src_id = 0x03002,
    30                                    .output_idx = 2,
    31                                 },
    32                                 {
    33                                    .src_id = 0x14002,
    34                                    .output_idx = 0,
    35                                 },
    36                                 {
    37                                    .src_id = 0x14003,
    38                                    .output_idx = 0,
    39                                 }
    40                               };
    41
    42 hexagon_nn_output fcOutputs[] = {
    43                                   {
    44                                     .rank = 4,
    45                                     .max_sizes = {1, 1, 1, 10, 0, 0, 0, 0},
    46                                     .elementsize = 4,
    47                                     .zero_offset = 2,
    48                                     .stepsize = 0.0
    49                                   },
    50                                   {
    51                                     .rank = 4,
    52                                     .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    53                                     .elementsize = 4,
    54                                     .zero_offset = 0,
    55                                     .stepsize = 0.0
    56                                   },
    57                                   {
    58                                     .rank = 4,
    59                                     .max_sizes = {1, 1, 1, 1, 0, 0, 0, 0},
    60                                     .elementsize = 4,
    61                                     .zero_offset = 0,
    62                                     .stepsize = 0.00000115
    63                                   }
    64                                 };
    65
    66 ret = hexagon_nn_append_node(nng_id, 0x04000, OP_QuantizedMatMul_8x8to32,
    67                               NN_PAD_NA, fcInputs, 6, fcOutputs, 3);
    Copy to clipboard

A bias node follows the fully connected node and accepts 6 inputs

> 
> 
> - [3] input activation, min and max from the output of the MatMul (fc) node
> - [3] biases, min and max added as constant nodes

and 3 outputs

> 
> 
> - [3] output activation, min and max

For QNN, the biases are added as a static tensor input to the fully connected (FC) node.
*See the QNN FC example below for details*.

**QNN FC**

![../../_static/resources/qnn_tutorial_migrate_fc_qnn.png](data:image/png;base64,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)

QNN fully connected node accepts 3 inputs:

> 
> 
> - [1] output activation from the convolution node
> - [1] weights static tensor
> - [1] biases static tensor

and 1 output

> 
> 
> - [1] output activation

QNN Fully Connected

1 // output (activation tensor)
     2 uint32_t fcOutputDimensions[] = {1, 10};
     3 Qnn_Tensor_t fcOutputTensor                                 = QNN_TENSOR_INIT;
     4 fcOutputTensor.version                                      = QNN_TENSOR_VERSION_2;
     5 fcOutputTensor.v2.id                                        = 0;
     6 fcOutputTensor.v2.name                                      = "fcOutput";
     7 fcOutputTensor.v2.type                                      = QNN_TENSOR_TYPE_NATIVE;
     8 fcOutputTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     9 fcOutputTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    10 fcOutputTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    11 fcOutputTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    12 fcOutputTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0024064707104117;
    13 fcOutputTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -151;
    14 fcOutputTensor.v2.rank                                      = 2;
    15 fcOutputTensor.v2.dimensions                                = fcOutputDimensions;
    16 fcOutputTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    17 fcOutputTensor.v2.clientBuf                                 = QNN_CLIENT_BUFFER_INIT;
    18 fcOutputTensor.v2.isDynamicDimensions                       = nullptr;
    19 fcOutputTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    20 fcOutputTensor.v2.isProduced                                = 0;
    21
    22 ret = QnnTensor_createGraphTensor(graph, &fcOutputTensor);
    23
    24 // weights (static tensor)
    25 static uint8_t fcWeights[] = {223, 128, 73, 137, 199, 155, ...};
    26 uint32_t fcWeightsDimensions[] = {10, 3600};
    27 Qnn_Tensor_t fcWeightsTensor                                 = QNN_TENSOR_INIT;
    28 fcWeightsTensor.version                                      = QNN_TENSOR_VERSION_2;
    29 fcWeightsTensor.v2.id                                        = 0;
    30 fcWeightsTensor.v2.name                                      = "fcWeights";
    31 fcWeightsTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
    32 fcWeightsTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    33 fcWeightsTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    34 fcWeightsTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    35 fcWeightsTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    36 fcWeightsTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0000707526414772;
    37 fcWeightsTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -128;
    38 fcWeightsTensor.v2.rank                                      = 2;
    39 fcWeightsTensor.v2.dimensions                                = fcWeightsDimensions;
    40 fcWeightsTensor.v2.clientBuf.data                            = fcWeights;
    41 fcWeightsTensor.v2.clientBuf.dataSize                        = 36000;
    42 fcWeightsTensor.v2.isDynamicDimensions                       = nullptr;
    43 fcWeightsTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    44 fcWeightsTensor.v2.isProduced                                = 0;
    45
    46 ret = QnnTensor_createGraphTensor(graph, &fcWeightsTensor);
    47
    48 // biases (static tensor)
    49 static uint8_t fcBiases[] = {230, 174, 41, 246, 26, 247, ...};
    50 uint32_t fcBiasesDimensions[] = {10};
    51 Qnn_Tensor_t fcBiasesTensor                                 = QNN_TENSOR_INIT;
    52 fcBiasesTensor.version                                      = QNN_TENSOR_VERSION_2;
    53 fcBiasesTensor.v2.id                                        = 0;
    54 fcBiasesTensor.v2.name                                      = "fcBiases";
    55 fcBiasesTensor.v2.type                                      = QNN_TENSOR_TYPE_STATIC;
    56 fcBiasesTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    57 fcBiasesTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    58 fcBiasesTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    59 fcBiasesTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    60 fcBiasesTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0000581612985115;
    61 fcBiasesTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -116;
    62 fcBiasesTensor.v2.rank                                      = 1;
    63 fcBiasesTensor.v2.dimensions                                = fcBiasesDimensions;
    64 fcBiasesTensor.v2.clientBuf.data                            = fcBiases;
    65 fcBiasesTensor.v2.clientBuf.dataSize                        = 10;
    66 fcBiasesTensor.v2.isDynamicDimensions                       = nullptr;
    67 fcBiasesTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    68 fcBiasesTensor.v2.isProduced                                = 0;
    69
    70 ret = QnnTensor_createGraphTensor(graph, &fcBiasesTensor);
    71
    72 // create a list of all input tensors
    73 Qnn_Tensor_t fcInputList[3] = {
    74                                 convOutputTensor, // output activation from convolution node
    75                                 fcWeightsTensor,
    76                                 fcBiasesTensor
    77                               };
    78
    79 // create a list of all output tensors
    80 Qnn_Tensor_t fcOutputList[1] = {fcOutputTensor};
    81
    82 // operation definition
    83 Qnn_OpConfig_t fcOpDefinition   = QNN_OPCONFIG_INIT;
    84 fcOpDefinition.v1.name          = "fc";
    85 fcOpDefinition.v1.packageName   = "qti.aisw";
    86 fcOpDefinition.v1.typeName      = QNN_OP_FULLY_CONNECTED;
    87 fcOpDefinition.v1.numOfParams   = 0;
    88 fcOpDefinition.v1.params        = nullptr;
    89 fcOpDefinition.v1.numOfInputs   = 3;
    90 fcOpDefinition.v1.inputTensors  = fcInputList;
    91 fcOpDefinition.v1.numOfOutputs  = 1;
    92 fcOpDefinition.v1.outputTensors = fcOutputList;
    93
    94 // add the node to the graph
    95 ret = QnnGraph_addNode(graph, fcOpDefinition);
    Copy to clipboard

### Add Nodes - SoftMax

**Hexagon-nn SoftMax**

![../../_static/resources/qnn_tutorial_migrate_softmax_hnn.png](data:image/png;base64,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)

Hexagon-nn fully SoftMax node accepts 3 inputs

> 
> 
> - [3] input activation from the output of the fully connected BiasAdd node

and 1 output

> 
> 
> - [1] output activation

Hexagon-nn SoftMax

1 hexagon_nn_input softMaxInputs[] = {
     2                                       {
     3                                          .src_id = (0x05000),
     4                                          .output_idx = (0),
     5                                       },
     6                                       {
     7                                          .src_id = (0x05001),
     8                                          .output_idx = (1),
     9                                       },
    10                                       {
    11                                          .src_id = (0x05002),
    12                                          .output_idx = (2),
    13                                       }
    14                                     };
    15
    16 hexagon_nn_output softMaxOutputs[] = {
    17                                         {
    18                                           .rank = 4,
    19                                           .max_sizes = {1, 1, 1, 10, 0, 0, 0, 0},
    20                                           .elementsize = 4,
    21                                           .zero_offset = 0,
    22                                           .stepsize = 0.0
    23                                         }
    24                                       };
    25
    26 ret = hexagon_nn_append_node(nng_id, 0x00024002, OP_Softmax_uint8,
    27                               NN_PAD_NA, softMaxInputs, 3, softMaxOutputs, 1);
    Copy to clipboard

**QNN SoftMax**

![../../_static/resources/qnn_tutorial_migrate_softmax_qnn.png](data:image/png;base64,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)

QNN SoftMax node accepts 1 input

> 
> 
> - [1] input activation from the output of the fully connected node

and 1 output

> 
> 
> - [1] output activation

QNN SoftMax

1 // output (activation tensor)
     2 uint32_t softMaxOutputDimensions[] = {1, 10};
     3 Qnn_Tensor_t softMaxOutputTensor                                 = QNN_TENSOR_INIT;
     4 softMaxOutputTensor.version                                      = QNN_TENSOR_VERSION_2;
     5 softMaxOutputTensor.v2.id                                        = 0;
     6 softMaxOutputTensor.v2.name                                      = "probOutput";
     7 softMaxOutputTensor.v2.type                                      = QNN_TENSOR_TYPE_APP_READ;
     8 softMaxOutputTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
     9 softMaxOutputTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    10 softMaxOutputTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    11 softMaxOutputTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    12 softMaxOutputTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.00390625;
    13 softMaxOutputTensor.v2.quantizeParams.scaleOffsetEncoding.offset = 0;
    14 softMaxOutputTensor.v2.rank                                      = 2;
    15 softMaxOutputTensor.v2.dimensions                                = softMaxOutputDimensions;
    16 softMaxOutputTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    17 softMaxOutputTensor.v2.clientBuf                                 = QNN_CLIENT_BUFFER_INIT;
    18 softMaxOutputTensor.v2.isDynamicDimensions                       = nullptr;
    19 softMaxOutputTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    20 softMaxOutputTensor.v2.isProduced                                = 0;
    21
    22 ret = QnnTensor_createGraphTensor(graph, &softMaxOutputTensor);
    23
    24 // create a list of all input tensors
    25 Qnn_Tensor_t softMaxInputList[1] = {fcOutputTensor};
    26
    27 // create a list of all output tensors
    28 Qnn_Tensor_t softMaxOutputList[1] = {softMaxOutputTensor};
    29
    30 // operation definition
    31 Qnn_OpConfig_t softMaxOpDefinition   = QNN_OPCONFIG_INIT;
    32 softMaxOpDefinition.v1.name          = "prob";
    33 softMaxOpDefinition.v1.packageName   = "qti.aisw";
    34 softMaxOpDefinition.v1.typeName      = QNN_OP_SOFTMAX;
    35 softMaxOpDefinition.v1.numOfParams   = 0;
    36 softMaxOpDefinition.v1.params        = nullptr;
    37 softMaxOpDefinition.v1.numOfInputs   = 1;
    38 softMaxOpDefinition.v1.inputTensors  = softMaxInputList;
    39 softMaxOpDefinition.v1.numOfOutputs  = 1;
    40 softMaxOpDefinition.v1.outputTensors = softMaxOutputList;
    41
    42 // add the node to the graph
    43 ret = QnnGraph_addNode(graph, softMaxOpDefinition);
    Copy to clipboard

### Add Nodes - output

**Hexagon-nn output**

![../../_static/resources/qnn_tutorial_migrate_output_hnn.png](data:image/png;base64,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)

Hexagon-nn requires an OUTPUT node to be added with inputs referring to tensors that will be the
graph’s output(s).  It will output tensors to the hexagon\_nn\_tensordef provided during the
call to hexagon\_nn\_execute. See the [Execute Graph](https://docs.qualcomm.com/doc/80-63442-50/topic/hexnn_migration.html#execute-graph) section below for more details.

Hexagon-nn Output

1 hexagon_nn_input outputInputs[] = {{
    2     .src_id = (0x06000),
    3     .output_idx = (0),
    4 }};
    5
    6 hexagon_nn_append_node(nng_id, 0x07000, OP_OUTPUT, NN_PAD_NA, outputInputs, 1, 0, 0);
    Copy to clipboard

**QNN Output**

Note

The QNN output is an activation tensor defined to be of type QNN\_TENSOR\_TYPE\_APP\_READ.
See QNN section in [Add Nodes - SoftMax](https://docs.qualcomm.com/doc/80-63442-50/topic/hexnn_migration.html#add-nodes-softmax) for example.

## Finalize Graph

Once the Hexagon-nn graph is constructed (above steps) it needs to be prepared (finalized) for
execution.

Hexagon-nn Finalize

1res = hexagon_nn_prepare(nng_id);
    Copy to clipboard

For QNN the graph’s input and output tensors need to be created and registered before finalizing the graph.
These tensors were registered when the QNN nodes were added to the graph.

Note

The QNN input is an activation tensor defined to be of type QNN\_TENSOR\_TYPE\_APP\_WRITE.
The QNN output is an activation tensor defined to be of type QNN\_TENSOR\_TYPE\_APP\_READ.

QNN Finalize

1 // don't forget to do all your error checking (ret == QNN_SUCCESS)
    2 Qnn_ProfileHandle_t profile{nullptr};
    3 Qnn_SignalHandle_t signal{nullptr};
    4 ret = QnnGraph_finalize(graph, profile, signal);
    Copy to clipboard

## Execute Graph

**Hexagon-nn**

The input and output tensor definitions need to be initialized before graph execution.

Hexagon-nn uses **hexagon\_nn\_tensordef** structures to describe the tensor attributes for the
inputs and outputs.

Hexagon-nn Execute

1 static float inputData    = { 15.0, 82.0, 85.0, 71.0, 20.0, ... };
     2 static float inputDataMin = { -0.003076 };
     3 static float inputDataMax = {  0.003076 };
     4 static int32_t inputDataNeedsQuantize [1] = { 1 };
     5
     6 hexagon_nn_tensordef in_tensor_defs[] =
     7   {
     8      {
     9         .batches = 1,
    10         .height = 32,
    11         .width = 32,
    12         .depth = 3,
    13         .data = (unsigned char *) inputData,
    14         .dataLen = 12288,
    15         .data_valid_len = 12288,
    16         .unused = 0,
    17      },
    18      {
    19         .batches = 1,
    20         .height = 1,
    21         .width = 1,
    22         .depth = 1,
    23         .data = (unsigned char *) inputDataMin,
    24         .dataLen = 4,
    25         .data_valid_len = 4,
    26         unused = 0,
    27      },
    28      {
    29         .batches = 1,
    30         .height = 1,
    31         .width = 1,
    32         .depth = 1,
    33         .data = (unsigned char *) inputDataMax,
    34         .dataLen = 4,
    35         .data_valid_len = 4,
    36         .unused = 0,
    37      },
    38      {
    39         .batches = 1,
    40         .height = 1,
    41         .width = 1,
    42         .depth = 1,
    43         .data = (unsigned char *) inputDataNeedsQuantize,
    44         .dataLen = 4,
    45         .data_valid_len = 4,
    46         .unused = 0,
    47      }
    48   };
    49
    50 float outData[10] = { };
    51
    52 hexagon_nn_tensordef out_tensor_defs[] = {
    53     {
    54         .batches = (1),
    55         .height = (1),
    56         .width = (1),
    57         .depth = (10),
    58         .data = (unsigned char *) outData,
    59         .dataLen = (40),
    60         .data_valid_len = (40),
    61         .unused = 0,
    62     }};
    63
    64  ret = hexagon_nn_execute_new(nng_id, in_tensor_defs, 4, out_tensor_defs, 1);
    Copy to clipboard

**QNN**

QNN uses the same **Qnn\_Tensor\_t** structures to describe the tensor attributes as well as
supply input activation data, and collect output activation data. Note that the tensors are
in quantized format to keep consistent with the input and output tensor types of the graph
created above.

QNN Execute

1 // Setup input buffer
     2 const uint32_t inputDataSize = 1 * 32 * 32 * 3 * sizeof(uint8_t);
     3 uint8_t *inputData = (uint8_t *)malloc(inputDataSize);
     4 memset(inputData, 0, inputDataSize);
     5
     6 // Fill in the quantized input tensor data
     7 ...
     8
     9 uint32_t graphInDimensions[] = {1, 32, 32, 3};
    10 Qnn_Tensor_t graphInTensor                                 = QNN_TENSOR_INIT;
    11 graphInTensor.version                                      = QNN_TENSOR_VERSION_2;
    12 graphInTensor.v2.id                                        = inputTensor.v2.id;
    13 graphInTensor.v2.name                                      = "graphIn";
    14 graphInTensor.v2.type                                      = QNN_TENSOR_TYPE_APP_WRITE;
    15 graphInTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    16 graphInTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    17 graphInTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    18 graphInTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    19 graphInTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.0256975870579481;
    20 graphInTensor.v2.quantizeParams.scaleOffsetEncoding.offset = -132;
    21 graphInTensor.v2.rank                                      = 4;
    22 graphInTensor.v2.dimensions                                = graphInDimensions;
    23 graphInTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    24 graphInTensor.v2.clientBuf.data                            = inputData;
    25 graphInTensor.v2.clientBuf.dataSize                        = inputDataSize;
    26 graphInTensor.v2.isDynamicDimensions                       = nullptr;
    27 graphInTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    28 graphInTensor.v2.isProduced                                = 0;
    29
    30 // Setup output buffer
    31 const uint32_t outputDataSize = 10 * sizeof(uint8_t);
    32 uint8_t *outputData = (uint8_t *)malloc(outputDataSize);
    33 memset(outputData, 0, outputDataSize);
    34
    35 uint32_t graphOutDimensions[] = {1, 10};
    36 Qnn_Tensor_t graphOutTensor                                 = QNN_TENSOR_INIT;
    37 graphOutTensor.version                                      = QNN_TENSOR_VERSION_2;
    38 graphOutTensor.v2.id                                        = softMaxOutputTensor.v2.id;
    39 graphOutTensor.v2.name                                      = "graphOut";
    40 graphOutTensor.v2.type                                      = QNN_TENSOR_TYPE_APP_READ;
    41 graphOutTensor.v2.dataFormat                                = QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER;
    42 graphOutTensor.v2.dataType                                  = QNN_DATATYPE_UFIXED_POINT_8;
    43 graphOutTensor.v2.quantizeParams.encodingDefinition         = QNN_DEFINITION_DEFINED;
    44 graphOutTensor.v2.quantizeParams.quantizationEncoding       = QNN_QUANTIZATION_ENCODING_SCALE_OFFSET;
    45 graphOutTensor.v2.quantizeParams.scaleOffsetEncoding.scale  = 0.00390625;
    46 graphOutTensor.v2.quantizeParams.scaleOffsetEncoding.offset = 0;
    47 graphOutTensor.v2.rank                                      = 2;
    48 graphOutTensor.v2.dimensions                                = graphOutDimensions;
    49 graphOutTensor.v2.memType                                   = QNN_TENSORMEMTYPE_RAW;
    50 graphOutTensor.v2.clientBuf.data                            = outputData;
    51 graphOutTensor.v2.clientBuf.dataSize                        = outputDataSize;
    52 graphOutTensor.v2.isDynamicDimensions                       = nullptr;
    53 graphOutTensor.v2.sparseParams                              = QNN_SPARSE_PARAMS_INIT;
    54 graphOutTensor.v2.isProduced                                = 0;
    55
    56 ret = QnnGraph_execute(graph, &graphInTensor, 1, &graphOutTensor, 1, profile, signal);
    57
    58 // Read the quantized output tensor data and dequantize the data if necessary
    59 ...
    Copy to clipboard

## Destroy Graph

The final step is to release all the resources associated with the graph.

Hexagon-nn Destroy Graph

1ret = hexagon_nn_teardown(nng_id);
    Copy to clipboard

For QNN the graph resources are associated with the context so freeing the context will free all
resources owned by the graph. Afterwards, device resources need to be released. Finally if the backend accelerator is not longer required, then it’s
resources can also be freed.

QNN Destroy Graph

1QnnContext_free(context, profile);
    2
    3QnnDevice_free(device);
    4
    5QnnBackend_free(backend);
    6
    7QnnLog_free(logger);
    Copy to clipboard

Last Published: Oct 10, 2025

[Previous Topic
Migrating](https://docs.qualcomm.com/bundle/publicresource/80-63442-50/topics/tutorials.md) [Next Topic
LoRA (Low Rank Adaptation)](https://docs.qualcomm.com/bundle/publicresource/80-63442-50/topics/tutorials.md)