当前位置：首页 > news >正文

onnx 1.16 doc学习笔记四：python API-If和Scan

news 2026/2/10 8:59:41

onnx作为一个通用格式，很少有中文教程，因此开一篇文章对onnx 1.16文档进行翻译与进一步解释，
onnx 1.16官方文档：https://onnx.ai/onnx/intro/index.html](https://onnx.ai/onnx/intro/index.html)，
如果觉得有收获，麻烦点赞收藏关注，目前仅在CSDN发布，本博客会分为多个章节，目前尚在连载中，详见专栏链接：
https://blog.csdn.net/qq_33345365/category_12581965.html

开始编辑时间：2024/2/21；最后编辑时间：2024/2/21

这是本教程的第四篇，其余内容见上述专栏链接。

ONNX with Python

本教程的第一篇：介绍了ONNX的基本概念。

在本教程的第二篇，介绍了ONNX关于Python的API，具体涉及一个简单的线性回归例子和序列化。

本教程的第三篇，包括python API的三个部分：初始化器Initializer；属性Attributes；算子集和元数据Opset和Metadata

在本篇将继续对更多的内容进行介绍。具体介绍子图相关内容。

子图：测试与循环
- If
- Scan

子图：测试与循环 Subgraph: test and loops

通常情况下，这些操作会被归类为控制流。一般来说，最好避免使用它们，因为它们不像矩阵操作那样高效，而且矩阵操作经过高度优化，速度更快。

If

可以使用 if 语句来实现一个test。它根据一个布尔值执行子图 A 或子图 B。这种用法不是很常见，因为一个函数通常需要批处理比较的结果。下面的示例代码：根据符号计算矩阵中所有浮点数的和，并返回 1 或 -1。

import numpy
import onnx
from onnx.helper import (make_node, make_graph, make_model, make_tensor_value_info)
from onnx.numpy_helper import from_array
from onnx.checker import check_model
from onnxruntime import InferenceSession# initializers
value = numpy.array([0], dtype=numpy.float32)
zero = from_array(value, name='zero')# Same as before, X is the input, Y is the output.
X = make_tensor_value_info('X', onnx.TensorProto.FLOAT, [None, None])
Y = make_tensor_value_info('Y', onnx.TensorProto.FLOAT, [None])# 第一个节点：对所有轴求和。
rsum = make_node('ReduceSum', ['X'], ['rsum'])
# 第二个节点：将rsum与0比较，输出为cond
cond = make_node('Greater', ['rsum', 'zero'], ['cond'])# 构建if正确时的图
# Input for then
then_out = make_tensor_value_info('then_out', onnx.TensorProto.FLOAT, None)
# 正确返回常数1
then_cst = from_array(numpy.array([1]).astype(numpy.float32))# If正确时的节点，并得到图
then_const_node = make_node('Constant', inputs=[],outputs=['then_out'],value=then_cst, name='cst1')then_body = make_graph([then_const_node], 'then_body', [], [then_out])# 构建If错误时的图
else_out = make_tensor_value_info('else_out', onnx.TensorProto.FLOAT, [5])
else_cst = from_array(numpy.array([-1]).astype(numpy.float32))else_const_node = make_node('Constant', inputs=[],outputs=['else_out'],value=else_cst, name='cst2')else_body = make_graph([else_const_node], 'else_body',[], [else_out])# 最后If节点把两个图当成属性
if_node = onnx.helper.make_node('If', ['cond'], ['Y'],then_branch=then_body, else_branch=else_body)# 最后的图，使用If节点和前两个节点。
graph = make_graph([rsum, cond, if_node], 'if', [X], [Y], [zero])
onnx_model = make_model(graph)
check_model(onnx_model)# 指定算子集
del onnx_model.opset_import[:]
opset = onnx_model.opset_import.add()
opset.domain = ''
opset.version = 15
onnx_model.ir_version = 8# 保存模型
with open("onnx_if_sign.onnx", "wb") as f:f.write(onnx_model.SerializeToString())# 输出
sess = InferenceSession(onnx_model.SerializeToString(),providers=["CPUExecutionProvider"])x = numpy.ones((3, 2), dtype=numpy.float32)
res = sess.run(None, {'X': x})print("result", res)
print()print(onnx_model)

这个代码的核心是构建if使用的两个图，然后用着两个图构建节点，参照输出进行进一步的了解：

result [array([1.], dtype=float32)]ir_version: 8
graph {node {input: "X"output: "rsum"op_type: "ReduceSum"}node {input: "rsum"input: "zero"output: "cond"op_type: "Greater"}node {input: "cond"output: "Y"op_type: "If"attribute {name: "else_branch"g {node {output: "else_out"name: "cst2"op_type: "Constant"attribute {name: "value"t {dims: 1data_type: 1raw_data: "\000\000\200\277"}type: TENSOR}}name: "else_body"output {name: "else_out"type {tensor_type {elem_type: 1shape {dim {dim_value: 5}}}}}}type: GRAPH}attribute {name: "then_branch"g {node {output: "then_out"name: "cst1"op_type: "Constant"attribute {name: "value"t {dims: 1data_type: 1raw_data: "\000\000\200?"}type: TENSOR}}name: "then_body"output {name: "then_out"type {tensor_type {elem_type: 1}}}}type: GRAPH}}name: "if"initializer {dims: 1data_type: 1name: "zero"raw_data: "\000\000\000\000"}input {name: "X"type {tensor_type {elem_type: 1shape {dim {}dim {}}}}}output {name: "Y"type {tensor_type {elem_type: 1shape {dim {}}}}}
}
opset_import {domain: ""version: 15
}

在这个代码里，“else”和“then”分支都非常简单，甚至可以用一个“Where”节点替换“If”节点，这样会更快。有趣的是，当两个分支都比较大，并且跳过一个分支效率更高时，情况就变得复杂了。

Scan

根据规格描述，Scan操作似乎有些复杂：将张量的一个维度循环遍历并将其结果存储在预分配张量中，这是一种很有用的方法。以下示例实现了经典的回归问题最近邻算法。第一步是计算输入特征 X 和训练集 W 之间的成对距离： $KaTeX parse error: Undefined control sequence: \norm at position 21: …X,W)=(M_{ij})=(\̲n̲o̲r̲m̲ ̲X_i-W^2_j)_{ij}$ . 然后使用一个 TopK 算子提取 k 个最近邻点。

注：在原教程里，此处的公式就是这样的，不保真

ONNX scan 算子可以用于迭代执行一系列操作，类似于 Python 中的 for 循环。

这个算子有些复杂，我会在更多学习后，给出更为简单的例子。

import numpy
from onnx import numpy_helper, TensorProto
from onnx.helper import (make_model, make_node, set_model_props, make_tensor, make_graph,make_tensor_value_info)
from onnx.checker import check_model# subgraph
initializers = []
nodes = []
inputs = []
outputs = []value = make_tensor_value_info('next_in', 1, [None, 4])
inputs.append(value)
value = make_tensor_value_info('next', 1, [None])
inputs.append(value)value = make_tensor_value_info('next_out', 1, [None, None])
outputs.append(value)
value = make_tensor_value_info('scan_out', 1, [None])
outputs.append(value)node = make_node('Identity', ['next_in'], ['next_out'],name='cdistd_17_Identity', domain='')
nodes.append(node)node = make_node('Sub', ['next_in', 'next'], ['cdistdf_17_C0'],name='cdistdf_17_Sub', domain='')
nodes.append(node)node = make_node('ReduceSumSquare', ['cdistdf_17_C0'], ['cdistdf_17_reduced0'],name='cdistdf_17_ReduceSumSquare', axes=[1], keepdims=0, domain='')
nodes.append(node)node = make_node('Identity', ['cdistdf_17_reduced0'],['scan_out'], name='cdistdf_17_Identity', domain='')
nodes.append(node)graph = make_graph(nodes, 'OnnxIdentity',inputs, outputs, initializers)# main graphinitializers = []
nodes = []
inputs = []
outputs = []opsets = {'': 15, 'ai.onnx.ml': 15}
target_opset = 15  # subgraphs# initializers
list_value = [23.29599822460675, -120.86516699239603, -144.70495899914215, -260.08772982740413,154.65272105889147, -122.23295157108991, 247.45232560871727, -182.83789715805776,-132.92727431421793, 147.48710175784703, 88.27761768038069, -14.87785569894749,111.71487894705504, 301.0518319089629, -29.64235742280055, -113.78493504731911,-204.41218591022718, 112.26561056133608, 66.04032954135549,-229.5428380626701, -33.549262642481615, -140.95737409864623, -87.8145187836131,-90.61397011283958, 57.185488100413366, 56.864151796743855, 77.09054590340892,-187.72501631246712, -42.779503579806025, -21.642642730674076, -44.58517761667535,78.56025104939847, -23.92423223842056, 234.9166231927213, -73.73512816431007,-10.150864499514297, -70.37105466673813, 65.5755688281476, 108.68676290979731, -78.36748960443065]
value = numpy.array(list_value, dtype=numpy.float64).reshape((2, 20))
tensor = numpy_helper.from_array(value, name='knny_ArrayFeatureExtractorcst')
initializers.append(tensor)list_value = [1.1394007205963135, -0.6848101019859314, -1.234825849533081, 0.4023416340351105,0.17742614448070526, 0.46278226375579834, -0.4017809331417084, -1.630198359489441,-0.5096521973609924, 0.7774903774261475, -0.4380742907524109, -1.2527953386306763,-1.0485529899597168, 1.950775384902954, -1.420017957687378, -1.7062702178955078,1.8675580024719238, -0.15135720372200012, -0.9772778749465942, 0.9500884413719177,-2.5529897212982178, -0.7421650290489197, 0.653618574142456, 0.8644362092018127,1.5327792167663574, 0.37816253304481506, 1.4693588018417358, 0.154947429895401,-0.6724604368209839, -1.7262825965881348, -0.35955315828323364, -0.8131462931632996,-0.8707971572875977, 0.056165341287851334, -0.5788496732711792, -0.3115525245666504,1.2302906513214111, -0.302302747964859, 1.202379822731018, -0.38732680678367615,2.269754648208618, -0.18718385696411133, -1.4543657302856445, 0.04575851559638977,-0.9072983860969543, 0.12898291647434235, 0.05194539576768875, 0.7290905714035034,1.4940791130065918, -0.8540957570075989, -0.2051582634449005, 0.3130677044391632,1.764052391052246, 2.2408931255340576, 0.40015721321105957, 0.978738009929657,0.06651721894741058, -0.3627411723136902, 0.30247190594673157, -0.6343221068382263,-0.5108051300048828, 0.4283318817615509, -1.18063223361969, -0.02818222902715206,-1.6138978004455566, 0.38690251111984253, -0.21274028718471527, -0.8954665660858154,0.7610377073287964, 0.3336743414402008, 0.12167501449584961, 0.44386324286460876,-0.10321885347366333, 1.4542734622955322, 0.4105985164642334, 0.14404356479644775,-0.8877857327461243, 0.15634897351264954, -1.980796456336975, -0.34791216254234314]
value = numpy.array(list_value, dtype=numpy.float32).reshape((20, 4))
tensor = numpy_helper.from_array(value, name='Sc_Scancst')
initializers.append(tensor)value = numpy.array([2], dtype=numpy.int64)
tensor = numpy_helper.from_array(value, name='To_TopKcst')
initializers.append(tensor)value = numpy.array([2, -1, 2], dtype=numpy.int64)
tensor = numpy_helper.from_array(value, name='knny_Reshapecst')
initializers.append(tensor)# inputs
value = make_tensor_value_info('input', 1, [None, 4])
inputs.append(value)# outputs
value = make_tensor_value_info('variable', 1, [None, 2])
outputs.append(value)# nodes
# 这里是scan算子
node = make_node('Scan', ['input', 'Sc_Scancst'], ['UU032UU', 'UU033UU'],name='Sc_Scan', body=graph, num_scan_inputs=1, domain='')
nodes.append(node)node = make_node('Transpose', ['UU033UU'], ['Tr_transposed0'],name='Tr_Transpose', perm=[1, 0], domain='')
nodes.append(node)node = make_node('Sqrt', ['Tr_transposed0'], ['Sq_Y0'],name='Sq_Sqrt', domain='')
nodes.append(node)node = make_node('TopK', ['Sq_Y0', 'To_TopKcst'], ['To_Values0', 'To_Indices1'],name='To_TopK', largest=0, sorted=1, domain='')
nodes.append(node)node = make_node('Flatten', ['To_Indices1'], ['knny_output0'],name='knny_Flatten', domain='')
nodes.append(node)node = make_node('ArrayFeatureExtractor',['knny_ArrayFeatureExtractorcst', 'knny_output0'], ['knny_Z0'],name='knny_ArrayFeatureExtractor', domain='ai.onnx.ml')
nodes.append(node)node = make_node('Reshape', ['knny_Z0', 'knny_Reshapecst'], ['knny_reshaped0'],name='knny_Reshape', allowzero=0, domain='')
nodes.append(node)node = make_node('Transpose', ['knny_reshaped0'], ['knny_transposed0'],name='knny_Transpose', perm=[1, 0, 2], domain='')
nodes.append(node)node = make_node('Cast', ['knny_transposed0'], ['Ca_output0'],name='Ca_Cast', to=TensorProto.FLOAT, domain='')
nodes.append(node)node = make_node('ReduceMean', ['Ca_output0'], ['variable'],name='Re_ReduceMean', axes=[2], keepdims=0, domain='')
nodes.append(node)# graph
graph = make_graph(nodes, 'KNN regressor', inputs, outputs, initializers)# model
onnx_model = make_model(graph)
onnx_model.ir_version = 8
onnx_model.producer_name = 'skl2onnx'
onnx_model.producer_version = ''
onnx_model.domain = 'ai.onnx'
onnx_model.model_version = 0
onnx_model.doc_string = ''
set_model_props(onnx_model, {})# opsets
del onnx_model.opset_import[:]
for dom, value in opsets.items():op_set = onnx_model.opset_import.add()op_set.domain = domop_set.version = valuecheck_model(onnx_model)
with open("knnr.onnx", "wb") as f:f.write(onnx_model.SerializeToString())print(onnx_model)

输出如下所示：

ir_version: 8
producer_name: "skl2onnx"
producer_version: ""
domain: "ai.onnx"
model_version: 0
doc_string: ""
graph {node {input: "input"input: "Sc_Scancst"output: "UU032UU"output: "UU033UU"name: "Sc_Scan"op_type: "Scan"attribute {name: "body"g {node {input: "next_in"output: "next_out"name: "cdistd_17_Identity"op_type: "Identity"domain: ""}node {input: "next_in"input: "next"output: "cdistdf_17_C0"name: "cdistdf_17_Sub"op_type: "Sub"domain: ""}node {input: "cdistdf_17_C0"output: "cdistdf_17_reduced0"name: "cdistdf_17_ReduceSumSquare"op_type: "ReduceSumSquare"attribute {name: "axes"ints: 1type: INTS}attribute {name: "keepdims"i: 0type: INT}domain: ""}node {input: "cdistdf_17_reduced0"output: "scan_out"name: "cdistdf_17_Identity"op_type: "Identity"domain: ""}name: "OnnxIdentity"input {name: "next_in"type {tensor_type {elem_type: 1shape {dim {}dim {dim_value: 4}}}}}input {name: "next"type {tensor_type {elem_type: 1shape {dim {}}}}}output {name: "next_out"type {tensor_type {elem_type: 1shape {dim {}dim {}}}}}output {name: "scan_out"type {tensor_type {elem_type: 1shape {dim {}}}}}}type: GRAPH}attribute {name: "num_scan_inputs"i: 1type: INT}domain: ""}node {input: "UU033UU"output: "Tr_transposed0"name: "Tr_Transpose"op_type: "Transpose"attribute {name: "perm"ints: 1ints: 0type: INTS}domain: ""}node {input: "Tr_transposed0"output: "Sq_Y0"name: "Sq_Sqrt"op_type: "Sqrt"domain: ""}node {input: "Sq_Y0"input: "To_TopKcst"output: "To_Values0"output: "To_Indices1"name: "To_TopK"op_type: "TopK"attribute {name: "largest"i: 0type: INT}attribute {name: "sorted"i: 1type: INT}domain: ""}node {input: "To_Indices1"output: "knny_output0"name: "knny_Flatten"op_type: "Flatten"domain: ""}node {input: "knny_ArrayFeatureExtractorcst"input: "knny_output0"output: "knny_Z0"name: "knny_ArrayFeatureExtractor"op_type: "ArrayFeatureExtractor"domain: "ai.onnx.ml"}node {input: "knny_Z0"input: "knny_Reshapecst"output: "knny_reshaped0"name: "knny_Reshape"op_type: "Reshape"attribute {name: "allowzero"i: 0type: INT}domain: ""}node {input: "knny_reshaped0"output: "knny_transposed0"name: "knny_Transpose"op_type: "Transpose"attribute {name: "perm"ints: 1ints: 0ints: 2type: INTS}domain: ""}node {input: "knny_transposed0"output: "Ca_output0"name: "Ca_Cast"op_type: "Cast"attribute {name: "to"i: 1type: INT}domain: ""}node {input: "Ca_output0"output: "variable"name: "Re_ReduceMean"op_type: "ReduceMean"attribute {name: "axes"ints: 2type: INTS}attribute {name: "keepdims"i: 0type: INT}domain: ""}name: "KNN regressor"initializer {dims: 2dims: 20data_type: 11name: "knny_ArrayFeatureExtractorcst"raw_data: ",\\&\212\306K7@\333z`\345^7^\300\304\312,\006\217\026b\300Z9dWgAp\300.+F\027\343Tc@\203\330\264\255\350\216^\300\260\022\216sy\356n@\237h\263\r\320\332f\300\224\277.;\254\235`\300\336\370lV\226ob@\261\201\362|\304\021V@c,[Mv\301-\300\322\214\240\223\300\355[@)\036\262M\324\320r@nE;\211q\244=\300\021n5`<r\\\300\207\211\201\2400\215i\300H\232p\303\377\020\\@\317K[\302\224\202P@&\306\355\355^\261l\300\301/\377<N\306@\300#w\001\317\242\236a\300$fd\023!\364U\300\204\327LIK\247V\300J\211\366\022\276\227L@\262\345\254\206\234nL@f{\013\201\313ES@\234\343hU3wg\300\3370\367\305\306cE\300\336A\347;\204\2445\300f\374\242\031\347JF\300\325\2557\'\333\243S@\331\354\345{\232\3547\300\307o)\372T]m@#\005\000W\014oR\300\'\025\227\034>M$\300\310\252\022\\\277\227Q\300l_\243\036\326dP@\333kk\354\363+[@\223)\036\363\204\227S\300"}initializer {dims: 20dims: 4data_type: 1name: "Sc_Scancst"raw_data: "\342\327\221?\267O/\277\306\016\236\277\271\377\315>3\2575>\314\361\354>;\266\315\276W\252\320\277\221x\002\277\234\tG?FK\340\276\231[\240\277\3746\206\277\002\263\371?&\303\265\277\020g\332\277$\014\357?b\375\032\276\342.z\277\3778s?/d#\300\207\376=\277\214S\'?\261K]?\0342\304?\205\236\301>\363\023\274?\212\252\036>^&,\277\324\366\334\277Z\027\270\276[*P\277\220\354^\277\241\rf=~/\024\277\320\203\237\276*z\235?m\307\232\276\225\347\231?\263O\306\276\251C\021@ \255?\276\250(\272\277Hm;=\265Dh\277\031\024\004>\262\304T=\256\245:?\374=\277?\005\246Z\277\002\025R\276iJ\240>x\314\341?\313j\017@h\341\314>\223\216z?.:\210=6\271\271\276\231\335\232>\357b\"\277 \304\002\277QN\333>\365\036\227\277k\336\346\2744\224\316\277\026\030\306>\227\330Y\276L=e\277^\323B?]\327\252>\3000\371=\013B\343>hd\323\275\242%\272?\3709\322>(\200\023>\355Ec\277\362\031 >\275\212\375\277\213!\262\276"}initializer {dims: 1data_type: 7name: "To_TopKcst"raw_data: "\002\000\000\000\000\000\000\000"}initializer {dims: 3data_type: 7name: "knny_Reshapecst"raw_data: "\002\000\000\000\000\000\000\000\377\377\377\377\377\377\377\377\002\000\000\000\000\000\000\000"}input {name: "input"type {tensor_type {elem_type: 1shape {dim {}dim {dim_value: 4}}}}}output {name: "variable"type {tensor_type {elem_type: 1shape {dim {}dim {dim_value: 2}}}}}
}
opset_import {domain: ""version: 15
}
opset_import {domain: "ai.onnx.ml"version: 15
}

示意图如下所示：

在这里插入图片描述

子图由操作符Scan执行。在本例中，有一个scan输入，这意味着操作符只构建一个输出。

node = make_node('Scan', ['X1', 'X2'], ['Y1', 'Y2'],name='Sc_Scan', body=graph, num_scan_inputs=1, domain='')

在第一步迭代中，子图接收 X1 和 X2 的第一行作为输入。它会产生两个输出：

第一个输出用于在下一轮迭代中替换 X1。
第二个输出存储在一个容器中，最终形成 Y2。

在第二步迭代中，子图的第二个输入是 X2 的第二行。

以下是用颜色简要说明这个过程：

绿色表示第一轮迭代。
蓝色表示第二轮迭代。

在这里插入图片描述

onnx 1.16 doc学习笔记四：python API-If和Scan

ONNX with Python

子图：测试与循环 Subgraph: test and loops

If

Scan

相关文章：

onnx 1.16 doc学习笔记四：python API-If和Scan

如何构建企业专属GPT

知识积累（二）：损失函数正则化与权重衰减

消息中间件-面试题

Python 将二维数组或矩阵变为三维

区块链与Solidity详细介绍及基本语法使用

题目 1253: 老王赛马

【MATLAB源码-第144期】基于matlab的蝴蝶优化算法（BOA）无人机三维路径规划，输出做短路径图和适应度曲线。

地下管线管网三维建模工具MagicPipe3D V3.4.2发布

糖尿病性视网膜病变（DR）的自动化检测和分期

C 标准库 - ＜errno.h＞

基于springboot+vue的房屋租赁管理系统(前后端分离)

Sora----打破虚实之间的最后一根枷锁----这扇门的背后是人类文明的晟阳还是最后的余晖

C语言之static关键字详解

Redis高性能原理

MSS与cwnd的关系，rwnd又是什么？

解决两个MySQL5.7报错

[OpenAI]继ChatGPT后发布的Sora模型原理与体验通道

机器人初识 —— 电机传动系统

安卓游戏开发之音频技术优劣分析

Leetcode 3576. Transform Array to All Equal Elements

UDP(Echoserver)

【第二十一章 SDIO接口(SDIO)】

【RockeMQ】第2节｜RocketMQ快速实战以及核⼼概念详解（二）

微信小程序云开发平台MySQL的连接方式

零基础在实践中学习网络安全-皮卡丘靶场（第九期-Unsafe Fileupload模块）（yakit方式）

用机器学习破解新能源领域的“弃风”难题

适应性Java用于现代 API：REST、GraphQL 和事件驱动

如何配置一个sql server使得其它用户可以通过excel odbc获取数据

【Linux】Linux安装并配置RabbitMQ