CN112506521B - Data stream model-oriented high-order calling code generation method and device - Google Patents
Data stream model-oriented high-order calling code generation method and device Download PDFInfo
- Publication number
- CN112506521B CN112506521B CN202011496767.XA CN202011496767A CN112506521B CN 112506521 B CN112506521 B CN 112506521B CN 202011496767 A CN202011496767 A CN 202011496767A CN 112506521 B CN112506521 B CN 112506521B
- Authority
- CN
- China
- Prior art keywords
- module
- calling
- calculation module
- condition
- order
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 101
- 238000004458 analytical method Methods 0.000 claims abstract description 40
- 239000011159 matrix material Substances 0.000 claims abstract description 36
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000007480 spreading Effects 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
- G06F8/42—Syntactic analysis
- G06F8/427—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
- G06F8/44—Encoding
- G06F8/447—Target code generation
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Stored Programmes (AREA)
Abstract
The invention discloses a data stream model-oriented high-order call code generation method and device. The method comprises the following steps: analyzing the calculation module in the data stream to obtain an analysis result of the calculation module; determining whether the calculation module is a submodule calling module according to the analysis result; determining the size relation between the array dimension corresponding to the calculation module and the preset maximum expansion times under the condition that the calculation module is a sub-module calling module or the calculation module is a non-sub-module calling module and the calculation module is a matrix function operation module; under the condition that the array dimension is larger than the maximum expansion times, performing function call based on a loop call mode to generate a high-order call code corresponding to the calculation module; and under the condition that the array dimension is smaller than the maximum expansion times, expanding a calling function according to the array dimension, and generating a high-order calling code corresponding to the computing module. The invention can simplify the generated codes and generate the concise codes which accord with the user to check.
Description
Technical Field
The invention relates to the technical field of code generation based on a model, in particular to a high-order calling code generation method and device for a data stream model.
Background
When a model is built, the common module calling condition is that the actual parameters are matched with the shape parameters. When the actual reference is not matched with the type of the shape parameter, if the actual reference is higher in order and meets the requirement that the higher order is unfolded to the lower order, the actual reference is still a legal model module call, which is defined as follows:
the types of the regulated modules are as follows: t n←m(T1,T2,...,Tn-1)
Wherein, the type of return value is indicated, and T 1,T2,...,Tn indicates the type of parameter.
Let the call expression be: r=m (e 1,e2,...,en-1)
The types of r and e 1,e2,...,en-1 are respectively: s n,S1,S2,...,Sn-1.
The following relationship holds:
this is considered a legal type of module call.
Wherein: t i＝Si represents the same type, T i＝deorder(Si, j) represents the same type after the reduction
Let t=int [3] [4], then deorder (T, 1) =int [4], deorder (T, 2) =int.
Namely, for a two-dimensional array, two order reduction operations are allowed to be carried out, and a one-dimensional array and a scalar are respectively obtained.
Code is generated in this calling mode, which includes two modes. In the generated codes, function call is sequentially carried out according to each specific value of the real parameter, when the dimension of the real parameter is smaller, the generated function call codes are simpler and have good readability, and when the dimension of the real parameter is larger, the generated function call codes are more, so that poor readability is caused; and in the code generated in the second mode, according to the dimension difference of the real reference shape parameter, performing function call in a cyclic call mode, wherein the time complexity and the space complexity of the code generated in the method are higher than those of the code generated in the first mode when the dimension of the real reference is small.
Disclosure of Invention
The invention solves the technical problems that: the method and the device for generating the high-order calling codes for the data stream model are provided for overcoming the defects of the prior art.
In order to solve the above technical problems, an embodiment of the present invention provides a method for generating a high-order call code for a data stream model, including:
Analyzing a calculation module in a data stream to obtain an analysis result of the calculation module;
Determining whether the computing module is a submodule calling module according to the analysis result;
Determining the size relation between the array dimension corresponding to the calculation module and the preset maximum expansion times under the condition that the calculation module is a sub-module calling module or the calculation module is a non-sub-module calling module and the calculation module is a matrix function operation module;
performing function call based on a loop call mode under the condition that the array dimension is larger than the maximum expansion times, and generating a high-order call code corresponding to the calculation module;
And under the condition that the array dimension is smaller than the maximum expansion times, expanding a calling function according to the array dimension, and generating a high-order calling code corresponding to the computing module.
Optionally, before the analyzing the computing module in the data stream to obtain the analysis result of the computing module, the method further includes:
And acquiring the maximum unfolding times set by service personnel.
Optionally, before determining the size relationship between the array dimension corresponding to the computing module and the preset maximum expansion times, the method further includes:
According to the analysis result, the shape parameter type of the calculation module and the real parameter type of the calculation module are obtained;
determining that the computing module is a one-dimensional high-order calling module under the condition that the shape parameter type is a basic data type and the real parameter type is a one-dimensional array;
And determining the computing module as a second-order high-order calling module under the condition that the shape parameter type is a one-dimensional array and the real parameter type is two-dimensional data.
Optionally, after determining whether the computing module is a submodule calling module according to the parsing result, the method further includes:
Under the condition that the computing module is a non-submodule calling module, determining whether the computing module is a matrix function operation module according to the analysis result;
Executing the step of determining the magnitude relation between the array dimension corresponding to the calculation module and the preset maximum expansion times under the condition that the calculation module is a matrix function operation module;
and under the condition that the computing module is a non-matrix function operation module, directly generating a module code corresponding to the computing module.
In order to solve the above technical problem, an embodiment of the present invention further provides a device for generating a high-order call code for a data stream model, including:
the analysis result acquisition module is used for analyzing the calculation module in the data stream to obtain an analysis result of the calculation module;
the calculation module determining module is used for determining whether the calculation module is a submodule calling module according to the analysis result;
the size relation determining module is used for determining the size relation between the array dimension corresponding to the computing module and the preset maximum expansion times under the condition that the computing module is a sub-module calling module or the computing module is a non-sub-module calling module and the computing module is a matrix function operation module;
The first high-order code generation module is used for carrying out function call based on a circular call mode under the condition that the array dimension is larger than the maximum expansion times, and generating a high-order call code corresponding to the calculation module;
And the second high-order code generation module is used for generating a high-order calling code corresponding to the calculation module according to the array dimension expansion calling function under the condition that the array dimension is smaller than the maximum expansion times.
Optionally, the apparatus further comprises:
And the maximum unfolding times acquisition module is used for acquiring the maximum unfolding times set by the service personnel.
Optionally, the apparatus further comprises:
The parameter type acquisition module is used for acquiring the shape parameter type of the calculation module and the real parameter type of the calculation module according to the analysis result;
The one-dimensional high-order call determining module is used for determining that the calculating module is a one-dimensional high-order call module under the condition that the shape parameter type is a basic data type and the real parameter type is a one-dimensional array;
The two-dimensional high-order modulation determining module is used for determining that the calculating module is a second-order high-order modulation module under the condition that the shape parameter type is a one-dimensional array and the real parameter type is two-dimensional data.
Optionally, the apparatus further comprises:
the matrix function operation determining module is used for determining whether the computing module is a matrix function operation module according to the analysis result under the condition that the computing module is a non-submodule calling module;
the size relation executing module is used for executing the size relation determining module under the condition that the calculating module is a matrix function operation module;
And the module code generation module is used for directly generating the module code corresponding to the calculation module under the condition that the calculation module is a non-matrix function operation module.
Compared with the prior art, the invention has the advantages that:
The method and the device for generating the high-order calling code facing the data stream model provide user configurable options on the basis of supporting the generation of the high-order calling code, and consider the efficiency, the readability and the flexibility of code generation. Analyzing a calculation module in a data stream to obtain an analysis result of the calculation module, determining whether the calculation module is a sub-module calling module according to the analysis result, determining a size relation between an array dimension corresponding to the calculation module and a preset maximum expansion number under the condition that the calculation module is the sub-module calling module or the calculation module is a non-sub-module calling module and the calculation module is a matrix function operation module, performing function calling based on a circular calling mode under the condition that the array dimension is larger than the maximum expansion number, generating a high-order calling code corresponding to the calculation module, and expanding a calling function according to the array dimension under the condition that the array dimension is smaller than the maximum expansion number, and generating the high-order calling code corresponding to the calculation module. According to the embodiment of the invention, when the matrix function operation code is generated, whether the matrix function operation code is generated by expanding is controlled according to the set code expansion times, so that the generated code is simplified; and supporting writing a pseudo code for high-order call, and controlling the code expansion according to the set code expansion maximum frequency boundary value to generate a concise code which accords with the view of a user.
Drawings
FIG. 1 is a flow chart of steps of a method for generating a high-order calling code for a data stream model according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of one dimension Gao Jiediao being undeployed by a code when the maximum number of development times is 3 according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of one-dimensional Gao Jiediao spreading by codes when the maximum spreading number is 6 according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a high-order call code generating device facing to a data stream model according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The technical idea of the embodiment of the application is as follows: when the function is called in a high order, whether the code is generated and expanded is controlled, and the method is also suitable for generating the code of matrix function operation, and when the code is generated and expanded according to the set maximum number of times of the code, whether the matrix function operation expands the code according to the dimension of the matrix is determined, so that the code with more conciseness and better readability is controlled and generated.
Example 1
Referring to fig. 1, a step flow chart of a high-order call code generating method for a data stream model according to an embodiment of the present invention is shown, and as shown in fig. 1, the method may specifically include the following steps:
Step 101: analyzing the calculation module in the data stream to obtain an analysis result of the calculation module.
The embodiment of the invention can be applied to a scene of automatically generating the high-order calling code of the computing module.
High order modulation: the incoming real reference function shape parameter is higher by first order
When a function is called and matrix functions are operated, a common calling mode is peer calling, namely, a function is defined as void foo (double in1, double in 2), and when the function is called, a calling code is double input1; double input2, foo1 (input 1, input 2), when the real parameters are the array double input1[3], double input2[3], it is a high order call, and similar high order call is shown in the following table 1:
Table 1:
in this embodiment, the maximum expansion number default_ FLATTEN _max may be preset by the service personnel. The specific value of the maximum expansion number may be determined according to the service requirement, which is not limited in this embodiment.
Firstly, the calculation module in the data stream can be analyzed to obtain an analysis result of the calculation module, namely, a grammar analysis result of the calculation module.
After the analysis result of the calculation module is obtained, step 102 is performed.
Step 102: and determining whether the computing module is a submodule calling module according to the analysis result.
After the analysis result of the calculation module is obtained, whether the calculation module is a submodule calling module or not can be determined according to the analysis result.
Before the step102, it may be determined whether the calculation module is a matrix function operation module according to the analysis result if the calculation module is determined to be a non-sub-module calling module, where the step of determining a size relationship between the array dimension corresponding to the calculation module and the preset maximum expansion number is performed if the calculation module is a matrix function operation module, and where the calculation module is determined to be a non-matrix function operation module, a module code corresponding to the calculation module is directly generated.
After determining whether the calculation module is a sub-module calling module according to the parsing result, step 103 is performed.
Step 103: and determining the size relation between the array dimension corresponding to the calculation module and the preset maximum expansion times under the condition that the calculation module is a sub-module calling module or the calculation module is a non-sub-module calling module and the calculation module is a matrix function operation module.
Under the condition that the calculation module is determined to be a sub-module calling module, the size relation between the array dimension corresponding to the calculation module and the preset maximum expansion times can be determined.
Or under the condition that the computing module is determined to be a non-submodule calling module and the computing module is a matrix function operation module, determining the size relation between the array dimension corresponding to the computing module and the preset maximum expansion times.
Of course, before the step 103, the shape parameter type of the calculation module and the real parameter type of the calculation module may be obtained according to the analysis result, where the calculation module is determined to be a one-dimensional high-order calling module when the shape parameter type is the basic data type and the real parameter type is a one-dimensional array. And under the condition that the shape parameter type is a one-dimensional array and the real parameter type is two-dimensional data, the determining and calculating module is a second-order high-order calling module.
In the event that it is determined that the array dimension of the computing module is greater than the maximum number of expansion, step 104 is performed.
And in the case where it is determined that the array dimension of the calculation module is smaller than the maximum expansion number, step 105 is performed.
Step 104: and under the condition that the array dimension is larger than the maximum expansion times, performing function call based on a loop call mode, and generating a high-order call code corresponding to the calculation module.
And under the condition that the array dimension of the calculation module is larger than the maximum expansion times, performing function call in a loop call mode to generate a high-order call code corresponding to the calculation module.
Step 105: and under the condition that the array dimension is smaller than the maximum expansion times, expanding a calling function according to the array dimension, and generating a high-order calling code corresponding to the computing module.
And under the condition that the array dimension of the calculation module is smaller than the maximum expansion times, expanding the calling function according to the array dimension of the calculation module to generate a high-order calling code corresponding to the calculation module.
Next, the implementation process of the present embodiment is described in detail below with reference to fig. 2 and 3.
As shown in fig. 2 and 3, the calculation module of the data flow implements a call to the function mlf, mlf is defined as double mlf (double val, double lmt), and a pseudo code for the function is generated, and the called pseudo code is written as shown in the following table 2:
table 2:
The treatment process is as follows:
1. obtaining the 4 th behavioral function call of the computing module through grammar analysis;
2. Analyzing the shape parameters of the function mlf, and determining the shape parameters of the function as basic data types;
3. Analyzing the real parameters of the function mlf, and determining the real parameters of the function as one-dimensional data;
4. Comparing the dimension of the real parameter one-dimensional array with the maximum function expansion times, if the maximum expansion times are set to 8, the generated pseudo codes need to be fully expanded, and the generated codes can be shown in the following table 3:
table 3:
if the maximum expansion times are set to 4, the function calling code is not expanded, and the function calling code is called according to a loop, and the generated code can be as shown in the following table 4:
table 4:
Example two
Referring to fig. 4, a schematic structural diagram of a high-order call code generating device for a data stream model according to an embodiment of the present invention is shown, and as shown in fig. 4, the device may specifically include the following modules:
The analysis result obtaining module 210 is configured to analyze a calculation module in a data stream to obtain an analysis result of the calculation module;
A calculation module determining module 220, configured to determine, according to the analysis result, whether the calculation module is a submodule calling module;
The size relation determining module 230 is configured to determine a size relation between an array dimension corresponding to the computing module and a preset maximum expansion number when the computing module is a sub-module calling module or the computing module is a non-sub-module calling module and the computing module is a matrix function operation module;
a first higher-order code generating module 240, configured to perform function call based on a loop call manner, and generate a higher-order call code corresponding to the calculating module, where the array dimension is greater than the maximum expansion number;
and the second high-order code generating module 250 is configured to generate a high-order calling code corresponding to the calculating module according to the array dimension expansion calling function when the array dimension is smaller than the maximum expansion times.
Optionally, the apparatus further comprises:
And the maximum unfolding times acquisition module is used for acquiring the maximum unfolding times set by the service personnel.
Optionally, the apparatus further comprises:
The parameter type acquisition module is used for acquiring the shape parameter type of the calculation module and the real parameter type of the calculation module according to the analysis result;
The one-dimensional high-order call determining module is used for determining that the calculating module is a one-dimensional high-order call module under the condition that the shape parameter type is a basic data type and the real parameter type is a one-dimensional array;
The two-dimensional high-order modulation determining module is used for determining that the calculating module is a second-order high-order modulation module under the condition that the shape parameter type is a one-dimensional array and the real parameter type is two-dimensional data.
Optionally, the apparatus further comprises:
the matrix function operation determining module is used for determining whether the computing module is a matrix function operation module according to the analysis result under the condition that the computing module is a non-submodule calling module;
the size relation executing module is used for executing the size relation determining module under the condition that the calculating module is a matrix function operation module;
And the module code generation module is used for directly generating the module code corresponding to the calculation module under the condition that the calculation module is a non-matrix function operation module.
What is not described in detail in the present specification is a well known technology to those skilled in the art.
Claims (4)
1. A method for generating a high-order call code for a data stream model, comprising:
Analyzing a calculation module in a data stream to obtain an analysis result of the calculation module, namely a grammar analysis result of the calculation module;
Determining whether the computing module is a submodule calling module according to the analysis result;
Under the condition that the computing module is a non-submodule calling module, determining whether the computing module is a matrix function operation module according to the analysis result; executing the step of determining the magnitude relation between the array dimension corresponding to the calculation module and the preset maximum expansion times under the condition that the calculation module is a matrix function operation module; under the condition that the computing module is a non-matrix function computing module, directly generating a module code corresponding to the computing module;
According to the analysis result, the shape parameter type of the calculation module and the real parameter type of the calculation module are obtained; determining that the computing module is a one-dimensional high-order calling module under the condition that the shape parameter type is a basic data type and the real parameter type is a one-dimensional array; determining the computing module as a second-order high-order calling module under the condition that the shape parameter type is a one-dimensional array and the real parameter type is two-dimensional data;
Determining the size relation between the array dimension corresponding to the calculation module and the preset maximum expansion times under the condition that the calculation module is a sub-module calling module or the calculation module is a non-sub-module calling module and the calculation module is a matrix function operation module;
performing function call based on a loop call mode under the condition that the array dimension is larger than the maximum expansion times, and generating a high-order call code corresponding to the calculation module;
And under the condition that the array dimension is smaller than the maximum expansion times, expanding a calling function according to the array dimension, and generating a high-order calling code corresponding to the computing module.
2. The method of claim 1, further comprising, prior to parsing the computing modules in the data stream to obtain parsing results for the computing modules:
And acquiring the maximum unfolding times set by service personnel.
3. A high order call code generation apparatus for a data stream model, comprising:
The analysis result acquisition module is used for analyzing the calculation module in the data stream to obtain an analysis result of the calculation module, namely a grammar analysis result of the calculation module;
the calculation module determining module is used for determining whether the calculation module is a submodule calling module according to the analysis result;
the matrix function operation determining module is used for determining whether the computing module is a matrix function operation module according to the analysis result under the condition that the computing module is a non-submodule calling module;
the size relation executing module is used for executing the size relation determining module under the condition that the calculating module is a matrix function operation module;
The module code generation module is used for directly generating the module code corresponding to the calculation module under the condition that the calculation module is a non-matrix function operation module;
The parameter type acquisition module is used for acquiring the shape parameter type of the calculation module and the real parameter type of the calculation module according to the analysis result;
The one-dimensional high-order call determining module is used for determining that the calculating module is a one-dimensional high-order call module under the condition that the shape parameter type is a basic data type and the real parameter type is a one-dimensional array;
The two-dimensional high-order call determining module is used for determining that the calculating module is a second-order high-order call module under the condition that the shape parameter type is a one-dimensional array and the real parameter type is two-dimensional data;
the size relation determining module is used for determining the size relation between the array dimension corresponding to the computing module and the preset maximum expansion times under the condition that the computing module is a sub-module calling module or the computing module is a non-sub-module calling module and the computing module is a matrix function operation module;
The first high-order code generation module is used for carrying out function call based on a circular call mode under the condition that the array dimension is larger than the maximum expansion times, and generating a high-order call code corresponding to the calculation module;
And the second high-order code generation module is used for generating a high-order calling code corresponding to the calculation module according to the array dimension expansion calling function under the condition that the array dimension is smaller than the maximum expansion times.
4. A device according to claim 3, characterized in that the device further comprises:
And the maximum unfolding times acquisition module is used for acquiring the maximum unfolding times set by the service personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011496767.XA CN112506521B (en) | 2020-12-17 | 2020-12-17 | Data stream model-oriented high-order calling code generation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011496767.XA CN112506521B (en) | 2020-12-17 | 2020-12-17 | Data stream model-oriented high-order calling code generation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112506521A CN112506521A (en) | 2021-03-16 |
| CN112506521B true CN112506521B (en) | 2024-05-14 |
Family
ID=74922115
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011496767.XA Active CN112506521B (en) | 2020-12-17 | 2020-12-17 | Data stream model-oriented high-order calling code generation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112506521B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IE920607A1 (en) * | 1991-02-27 | 1992-09-09 | Digital Equipment Corp | Method of constructing a constant-folding mechanism ini a¹multilanguage optimizing compiler |
| US9417853B1 (en) * | 2014-10-27 | 2016-08-16 | Dspace Digital Signal Processing And Control Engineering Gmbh | Method for generating a code for an electronic control unit |
| CN110309631A (en) * | 2019-07-12 | 2019-10-08 | 北京智游网安科技有限公司 | A kind of programming language structure obscures processing method, intelligent terminal and storage medium |
| CN111324355A (en) * | 2020-02-11 | 2020-06-23 | 苏州浪潮智能科技有限公司 | Method and system for debugging many-core code |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7099812B2 (en) * | 1999-09-24 | 2006-08-29 | Intrinsity, Inc. | Grid that tracks the occurrence of a N-dimensional matrix of combinatorial events in a simulation using a linear index |
| US7320121B2 (en) * | 2002-08-01 | 2008-01-15 | Sas Institute Inc. | Computer-implemented system and method for generating embedded code to add functionality to a user application |
| US8739137B2 (en) * | 2006-10-19 | 2014-05-27 | Purdue Research Foundation | Automatic derivative method for a computer programming language |
| US8060857B2 (en) * | 2009-01-31 | 2011-11-15 | Ted J. Biggerstaff | Automated partitioning of a computation for parallel or other high capability architecture |
| US9712936B2 (en) * | 2015-02-03 | 2017-07-18 | Qualcomm Incorporated | Coding higher-order ambisonic audio data with motion stabilization |
-
2020
- 2020-12-17 CN CN202011496767.XA patent/CN112506521B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IE920607A1 (en) * | 1991-02-27 | 1992-09-09 | Digital Equipment Corp | Method of constructing a constant-folding mechanism ini a¹multilanguage optimizing compiler |
| US9417853B1 (en) * | 2014-10-27 | 2016-08-16 | Dspace Digital Signal Processing And Control Engineering Gmbh | Method for generating a code for an electronic control unit |
| CN110309631A (en) * | 2019-07-12 | 2019-10-08 | 北京智游网安科技有限公司 | A kind of programming language structure obscures processing method, intelligent terminal and storage medium |
| CN111324355A (en) * | 2020-02-11 | 2020-06-23 | 苏州浪潮智能科技有限公司 | Method and system for debugging many-core code |
Non-Patent Citations (4)
| Title |
|---|
| FPGA仿真验证工具及前沿技术综述;赵欢等;《电子测试》;20200715(第13期);100-103 * |
| LIFT: A functional data-parallel IR for high-performance GPU code generation;Michel Steuwer等;《2017 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)》;20170228;74-85 * |
| 数组维度类型程序设计方法及高性能FFT实现;崔翔等;《软件学报》;20151231(第12期);3104-3116 * |
| 飞机机载数据滤波融合与聚类侦测算法研究;戴周云;《中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）》;20200115(第01期);C031-192 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112506521A (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110704518B (en) | Business data processing method and device based on Flink engine | |
| CN108984155B (en) | Data processing flow setting method and device | |
| CN111462728A (en) | Method, apparatus, electronic device and computer readable medium for generating speech | |
| CN111026368B (en) | Python-based plug-in generation method, device, equipment and storage medium | |
| CN115878096A (en) | Deep learning model unified application method, device, server and storage medium | |
| CN113885845B (en) | Calculation map generation method, system, equipment and medium of deep learning compiler | |
| CN112506521B (en) | Data stream model-oriented high-order calling code generation method and device | |
| CN113296760A (en) | Method for generating model code, computer device and readable storage medium | |
| CN112084735A (en) | FPGA (field programmable Gate array) cutting method and system based on RTL (real time language) source code | |
| US8954307B1 (en) | Chained programming language preprocessors for circuit simulation | |
| CN116360787A (en) | Automatic generation method and device for form codes, electronic equipment and storage medium | |
| CN110895460A (en) | Jenkins-based robot system integration method and device and terminal equipment | |
| CN113805966A (en) | Multi-terminal device dynamic view rendering implementation method and system | |
| CN112287396A (en) | Data processing method and device based on privacy protection | |
| Lim et al. | Architecture Exploration and Customization Tool of Deep Neural Networks for Edge Devices | |
| CN112613614A (en) | Method, system, equipment and medium for compressing and accelerating bert model | |
| CN106776322B (en) | method for dynamically configuring, loading and operating test cases | |
| CN112183768B (en) | Intelligent deep learning process assembling method based on semantic net | |
| Hodges et al. | Accelerating wireless intelligent network standards through formal techniques | |
| CN109165012B (en) | Non-interpretation type micro application implementation method and system | |
| CN115826961B (en) | Design method and device for data analysis system for separated reuse of designer and engine | |
| CN111600902B (en) | Data analysis platform with automatic data processing and multi-protocol access | |
| Yepes-Arbos et al. | I/O scalability boost for the next generation of Earth system models: IFS-XIOS integration as a case study | |
| CN117008891A (en) | Automatic debugging system and method, computer equipment and storage medium suitable for FPGA prototype verification | |
| CN116225398A (en) | Image processing method, device, processor and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |