Information generation apparatus, information generation method, and program

Abstract

To efficiently generate circuit symbol data and a design rule check parameter.SOLUTION: An information generation apparatus includes: analysis means which acquires text information of a data sheet and layout information which is obtained by partitioning the region of the data sheet into sections; extraction means which obtains extraction information, which is a value of an item necessary for generating a table sheet that stores information for use in generating circuit symbol data and a design rule check parameter, the extraction information being output from a machine-learned model that receives, as input, the data sheet, the text information, and the layout information; input-output category name acquisition means which acquires, from an input-output classification sheet, using the extraction information, an input category name and an output category name corresponding to the item of the extraction information, in association with the extraction information; and table sheet generation means which generates a table sheet using the extraction information and the input category name and the output category name associated with the extraction information.SELECTED DRAWING: Figure 11

Description

[Technical field]

[0001] The present disclosure relates to an information generating device, an information generating method, and a program. [Background technology]

[0002] Electric circuits are designed using a CAD (Computer-Aided Design) device. A user designs electric circuits using circuit symbol data registered in the CAD device.

[0003] The processing device described in Patent Document 1 generates circuit symbol data based on a pin map database including pin information. The processing device converts the generated circuit symbol data into circuit symbol data in a file format as a library data format of a CAD device. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2000-322463 A Summary of the Invention [Problem to be solved by the invention]

[0005] Some CAD devices have a design rule check (DRC) function that checks whether a circuit designed according to design rules is designed correctly. When DRC is executed, DRC parameters are set, which are indices for determining which of multiple design rules to use for checking. When generating DRC parameters, the information generating device needs to read library data of circuit symbols and obtain pin information. In other words, when creating a library for electric circuit CAD, it is necessary to obtain pin information twice, which is inefficient.

[0006] An object of the present disclosure is to provide an information generating device, an information generating method, and a program that enable efficient generation of circuit symbol data and DRC parameters. [Means for solving the problem]

[0007] An information generating device of one embodiment of the present disclosure includes an analysis means for acquiring text information of a data sheet and layout information, which is information dividing an area of ​​a data sheet into sections, from a data sheet for each component used in designing an electrical circuit; an extraction means for inputting the data sheet, the text information, and the layout information into a machine-learned learned model and acquiring, as an output, extracted information, which is values ​​of items necessary for generating a table sheet that stores circuit symbol data and information used to generate design rule check parameters; an input / output classification name acquisition means for using the extracted information to acquire input classification names and output classification names corresponding to items of the extracted information by linking them to the extracted information from an input / output classification sheet, which is a sheet used to search for input classification names and output classification names, which are items necessary for generating design rule check parameters; and a table sheet generation means for generating a table sheet using the extracted information and the input classification names and output classification names linked to the extracted information.

[0008] In one embodiment of the information generation method of the present disclosure, a computer acquires text information of the data sheet and layout information, which is information dividing the areas of the data sheet into sections, from a data sheet for each component used in designing an electrical circuit, inputs the data sheet, text information, and layout information into a machine-learned trained model, and obtains, as output from the trained model, extraction information, which is the values ​​of items necessary for generating a table sheet that stores circuit symbol data and information used to generate design rule check parameters, and uses the extraction information to obtain input classification names and output classification names corresponding to the items of the extraction information by linking them to the extraction information from an input / output classification sheet, which is a sheet used to search for input classification names and output classification names, which are items necessary for generating design rule check parameters, and generates a table sheet using the extraction information and the input classification name and output classification name linked to the extraction information.

[0009] A program according to one embodiment of the present disclosure causes a computer to execute the following processes: acquiring text information from a data sheet for each component used in designing an electrical circuit, and layout information, which is information dividing an area of ​​the data sheet into sections; inputting the data sheet, text information, and layout information into a machine-learned, trained model; acquiring, as output from the trained model, extraction information, which is values ​​of items necessary for generating a table sheet that stores circuit symbol data and information used to generate design rule check parameters; using the extraction information to acquire input classification names and output classification names corresponding to items of the extraction information by linking them to the extraction information from an input / output classification sheet, which is a sheet used to search for input classification names and output classification names, which are items necessary for generating design rule check parameters; and generating a table sheet using the extraction information and the input classification name and output classification name linked to the extraction information. Effect of the Invention

[0010] According to the present disclosure, it is possible to provide an information generating device, an information generating method, and a program that enable efficient generation of circuit symbol data and DRC parameters. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram showing an example of a configuration of an information generation system according to the present disclosure. [Diagram 2] FIG. 1 is a diagram showing an example of a data sheet according to the present disclosure. [Diagram 3] FIG. 13 is a diagram illustrating an example of an input classification sheet according to the present disclosure. [Figure 4] FIG. 13 is a diagram showing an example of an output classification sheet according to the present disclosure. [Diagram 5] FIG. 13 is a diagram showing an example of teacher data of a trained model used by an extraction unit according to the present disclosure. [Figure 6] FIG. 13 is a diagram showing an example of teacher data of a trained model used by an extraction unit according to the present disclosure. [Figure 7] FIG. 2 is a diagram showing an example of a block diagram of information included in a data sheet according to the present disclosure. [Figure 8] FIG. 13 is a diagram illustrating an example of an input request screen according to the present disclosure. [Figure 9] FIG. 1 is a diagram showing an example of a table sheet according to the present disclosure. [Figure 10] 10 is a flowchart illustrating an example of an operation of the information generating device according to the present disclosure. [Figure 11] FIG. 1 is a block diagram showing an example of a configuration of an information generating device according to the present disclosure. [Figure 12] 10 is a flowchart illustrating an example of an operation of the information generating device according to the present disclosure. [Figure 13] FIG. 2 is a diagram illustrating an example of a hardware configuration for executing control and processing according to each embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. The embodiments described below are limited in a manner that is technically preferable for implementing the present disclosure, but the scope of the invention is not limited to the following. In all drawings used to explain the following embodiments, the same reference numerals are used for similar parts unless otherwise specified. In the following embodiments, repeated explanations of similar configurations and operations may be omitted.

[0013] (First embodiment) First, an information generating system according to a first embodiment will be described with reference to the drawings. The information generating system of this embodiment is used in the design of an electric circuit using a CAD (Computer-Aided Design) device. The information generating system of this embodiment generates circuit symbol data and DRC (Design Rule Check) parameters used in the design of the electric circuit. The information generating system of this embodiment can be applied to the design of any electric circuit, not limited to the design of an electric circuit using a CAD device.

[0014] (Information Generation System) 1 is a diagram showing an example of a configuration of an information generation system according to the present disclosure. The information generation system 1 includes an information generation device 10, a data sheet storage device 110, and an input / output classification sheet storage device 140. The data sheet storage device 110 and the input / output classification sheet storage device 140 may be configured as a single storage device.

[0015] The data sheet storage device 110 stores a plurality of data sheets 70. The data sheet 70 is a document provided by a manufacturer of an electrical component. For example, the data sheet 70 includes information such as package dimensions, pin arrangement, pin functions, absolute maximum specifications, and electrical characteristics. The data sheet 70 is prepared for each component. Therefore, the data sheet 70 includes a data sheet 70 for each component required for designing an electrical circuit. When the number of components required for design is n, n data sheets 70-1 to 70-n corresponding to each of the n components are prepared (n is a natural number). In the following description, when it is not necessary to distinguish between the individual data sheets, they are referred to as data sheets 70. The contents of the data sheet 70 stored in the data sheet storage device 110 are broadly divided into those in a Portable Document Format (PDF) format including text information and layout information, and those in a format consisting of only image information. The layout information is information obtained by dividing the area of ​​the data sheet 70 into sections based on the contents of the data sheet 70. For example, the data sheet 70 is divided into sections by using the chapters or sections of the data sheet 70 as units. Each section is assigned a section name, for example, the headings contained in that section.

[0016] The information included in the data sheet 70 will be further described with reference to FIG. 2. FIG. 2 is a diagram showing an example of a data sheet. FIG. 2 shows a data sheet 70-1 for a part C (ABCD1234). In the example of FIG. 2, the data sheet 70-1 is divided into sections S1 to S5. The heading of each section is set as the section name. Section S1 is information showing the characteristics of the part C, and the section name is "Feature". Section S2 is information showing the pin arrangement, and the section name is "Pin Configuration". Section S3 is information showing the pin function, and the section name is "Pin Description". Section S4 is information showing the electrical characteristics, and the section name is "Electrical Characteristics". Section S5 is information showing the block diagram, and the section name is "Block Diagram". The information shown in FIG. 2 is an example, and the data sheet 70-n may include information other than the above. Furthermore, the information included in each section differs for each data sheet.

[0017] The input / output classification sheet storage device 140 stores an input / output classification sheet 80. The input / output classification sheet 80 is a sheet used to search for input classification names and output classification names required for generating DRC parameters. The input / output classification sheet 80 is prepared in advance separately from the data sheet 70. The input / output classification sheet 80 has an input classification sheet 81 and an output classification sheet 82. FIG. 3 is a diagram showing an example of the input classification sheet 81 (input classification sheet 810). The input classification sheet 810 of FIG. 3 includes information on items such as the power supply voltage of the interface (I / F), the interface (I / F) type, the input classification name, the main IC name, and the design specifications (IC: Integrated Circuit). FIG. 4 is a diagram showing an example of the output classification sheet 82 (output classification sheet 820). The output classification sheet 820 of FIG. 4 includes information on items such as the power supply voltage, type, output classification name, the main IC name, and the design specifications of the circuit.

[0018] The information processing terminal 60 is an information processing terminal used by a designer. The information processing terminal 60 includes a display unit (not shown) that displays information received from the information generating device 10, and an input unit (not shown) for transmitting information to the information generating device 10. For example, the display unit is realized by a display or the like. For example, the input unit is realized by a keyboard or a touch panel.

[0019] (Information generation device) Next, the information generating device 10 will be described with reference to Fig. 1. The information generating device 10 includes an analysis unit 11, an extraction unit 12, an input request unit 13, an input / output classification name acquisition unit 14, a table sheet generation unit 15, a circuit symbol generation unit 16, a DRC parameter generation unit 17, and a storage unit 18.

[0020] The analysis unit 11 obtains the text information and layout information of the data sheet 70 .

[0021] When the content of the data sheet 70 is in the PDF format including text information, layout information, and the like, the analysis unit 11 acquires the text information and layout information incorporated in the PDF.

[0022] When the content of the data sheet 70 is in a format consisting of only image information, the analysis unit 11 extracts text information from the image information using optical character recognition technology (OCR: Optical Character Reader). The analysis unit 11 identifies layout information by analyzing the extracted text information using a known natural language processing technology. Alternatively, the analysis unit 11 identifies layout information by analyzing the image information using a known image processing technology.

[0023] An example of a means for identifying layout information by a known natural language processing technique will be described below. For example, the analysis unit 11 may divide the data sheet 70 into sections by identifying numbers (such as 1-1) that are assumed to be chapter numbers or section numbers from the text information. In this case, the analysis unit 11 sets the sentence immediately following the chapter number or section number as the section name. As another example, the analysis unit 11 may identify a heading from the text information of the data sheet 70, thereby identifying the heading. For example, the analysis unit 11 identifies a heading from the text information of the data sheet 70 by using a list of headings stored in advance. In this case, the analysis unit 11 sets the identified heading as the name of the section. The above-mentioned means are only an example, and the analysis unit 11 may identify layout information by other means.

[0024] An example of a means for identifying layout information by a known image processing technique will be described below. For example, the analysis unit 11 may divide the data sheet 70 into sections by using the width of the margins. Alternatively, the analysis unit 11 may identify the headings of the data sheet 70 by using the font size, thereby dividing the data sheet into sections. The analysis unit 11 sets the identified headings as the names of the sections. The above-mentioned means are merely examples, and the analysis unit 11 may identify the layout information by other means.

[0025] When a section includes a table, the analysis unit 11 uses the text information to identify the table items and the order of arrangement of the items. For example, in the table section S3 in FIG. 2, the table items are "Number|Name|Type|Description" in the order shown on the left. The analysis unit 11 uses the text information to identify the table items and the order of arrangement of the items as table information, and stores it in the storage unit 18.

[0026] The extraction unit 12 extracts values ​​of items necessary for generating a table sheet from the data sheet 70. The extraction unit 12 inputs the data sheet 70 and the text information, layout information, and table information identified by the analysis unit 11 to a machine-learned model, and obtains values ​​of items necessary for generating a table sheet as an output.

[0027] The trained model that has been machine-learned will be described with reference to FIG. 5 and FIG. 6. FIG. 5 and FIG. 6 are diagrams showing an example of training data of the trained model. FIG. 5 is a diagram showing an example of training data for extracting information from a table. In the example of FIG. 5, the training data includes a search target, a query, a section name, and an extraction item. The search target is information indicating which item of the table sheet the extracted information is to be reflected in. The query is a keyword or a key sentence used when extracting information. The section name is a section name in which the information to be extracted is stored. The extraction item is an item of information to be extracted from the table information. FIG. 6 is a diagram showing an example of training data for extracting input / output (I / O) information from the direction of an arrow in a block diagram. The training data includes a search symbol and a value. The search symbol is information indicating the direction of the arrow. The value is information indicating the type of input / output (I / O) corresponding to the direction of the arrow. The trained model may be one model trained using both FIG. 5 and FIG. 6 as a training model, or may be two models trained using each of FIG. 5 and FIG. 6 as a training model.

[0028] An example of the extraction unit 12 extracting information using a trained model will be described with reference to FIG. 2. The extraction unit 12 inputs the data sheet 70 of FIG. 2 and the text information, layout information, and table information identified by the analysis unit 11 to the trained model trained using the information shown in FIG. 5 as training data. The trained model identifies an area including a query from an area of ​​a section name. For example, the trained model identifies an area including a query "VIN", that is, the first row of a table, from the section name "Electrical Characteristics" (section S4) in FIG. 2. The trained model outputs an extraction item as a value to be searched from an area including a query. For example, the trained model outputs the value "Max." as a value of the terminal withstand voltage from the first row of the table identified as an area including a query in FIG. 2. The extraction unit 12 extracts the value output from the trained model as a value to be input to a table sheet. The extraction unit 12 stores the extracted information in the storage unit 18 as extracted information.

[0029] An example of the extraction unit 12 extracting information using the trained model will be further described with reference to FIG. 7. FIG. 7 is a diagram showing an example of information in section S5 in the data sheet 70-1 in FIG. 2. In FIG. 7, information in the area surrounded by a dashed line is omitted. The extraction unit 12 inputs the data sheet 70 in FIG. 2 (FIG. 7) and the text information, layout information, and table information specified by the analysis unit 11 to the trained model trained using the information shown in FIG. 6 as training data. The trained model specifies an area including an arrow, the direction of the arrow, and the name of the pin at the end of the arrow by image processing from the area of ​​the section name "Block Diagram". For example, the trained model specifies an arrow in the upper left, that the arrow direction is leftward, and the pin name "OE1" at the end of the arrow from the area of ​​the section name "Block Diagram" in FIG. 7. The trained model outputs information indicating the type of input / output (I / O) of the identified pin. For example, the trained model outputs that the type of input / output of the pin "OE1" in FIG. 7 is I (Input). The extraction unit 12 extracts values ​​output from the trained model as values ​​to be input to the table sheet. The extraction unit 12 stores the extracted information in the storage unit 18 as extracted information.

[0030] When extracting information, the trained model may also output a value indicating the reliability of the extracted information. The reliability is a value that quantifies the likelihood of the value output by the trained model, and is calculated using a known method. For example, the reliability is a value that quantifies the similarity between the training data and the data sheet 70. The reliability is linked to the extracted information and stored in the storage unit 18.

[0031] The input request unit 13 requests an input related to a change in the extracted information from the information processing terminal 60. The input request unit 13 outputs the extracted information to the information processing terminal 60 and requests an input related to the change in the information. The input request unit 13 acquires the information related to the change transmitted from the information processing terminal 60.

[0032] A specific example in which the input request unit 13 requests an input related to a change in the extracted information will be described with reference to FIG. 8. FIG. 8 is a diagram showing an example of an input request screen. First, the input request unit 13 outputs an input request screen including the extracted information to the information processing terminal 60. For example, the input request screen includes the extracted information (D1) and a GUI (Graphical User Interface) (B1) for receiving an input related to a change in the information. The input request unit 13 may highlight a portion for which input is requested on the input request screen. For example, in the example shown in FIG. 6, the color of a field in which a value is blank is changed. The input request unit 13 may also highlight values ​​extracted by the extraction unit 12 that have low reliability. For example, in the example shown in FIG. 8, if the fields with reliability equal to or lower than a predetermined value are No. 1 "Terminal Withstand Voltage" and "I / F Type", the input request unit 13 changes the color of these fields. The information processing terminal 60 edits the extracted information D1 displayed on the input request screen via the input unit. When the information processing terminal 60 detects that the input has been completed, it transmits the input information to the information generating device 10. For example, the information processing terminal 60 transmits the input information to the information generating device 10 in response to detection of pressing of the "Confirm" button B2. The input request unit 13 receives the transmitted information. The input request unit 13 uses the received information to change the extracted information stored in the storage unit 18. Note that the trained model used by the extraction unit 12 may be re-trained using the information related to the change received from the input request unit 13. Re-training refers to adjusting the trained model by providing new teacher data. Re-training can improve the judgment accuracy of the trained model.

[0033] The input / output classification name acquisition unit 14 acquires an input classification name and an output classification name. The input / output classification name acquisition unit 14 searches the input / output classification sheet 80 using the values ​​of each item included in the extraction information stored in the storage unit 18. If there is an item on the input / output classification sheet that matches the extraction information, the input / output classification name acquisition unit 14 acquires the input classification name or output classification name corresponding to that item by linking it to the extraction information.

[0034] The table sheet generating unit 15 generates a table sheet using the extraction information stored in the storage unit 18 and the input classification name and output classification name associated with the extraction information. FIG. 9 is a diagram showing an example of a table sheet (table sheet 150). The table sheet generating unit 15 generates a table sheet by combining the extraction information (D2) with the input classification name and the output classification name (D3). Items of the generated table sheet are classified into a common part, a part dedicated to circuit symbol generation, and a part dedicated to DRC parameter generation. Items included in the common part are used for generating circuit symbol data and DRC parameters. Items included in the part dedicated to circuit symbol generation are used for generating circuit symbol data. Items included in the part dedicated to DRC parameter generation are used for generating DRC parameters. The table sheet generating unit 15 stores the generated table sheet in the storage unit 18. In addition, the table sheet generating unit 15 outputs the generated table sheet to the information processing terminal 60.

[0035] The circuit symbol generating unit 16 uses the generated table sheet to generate circuit symbol data in a predetermined format. The circuit symbol generating unit 16 generates the circuit symbol data using data in the common portion and the portion dedicated to circuit symbol generation of the table sheet. The circuit symbol generating unit 16 generates the circuit symbol data in accordance with a predetermined format of the CAD device. The circuit symbol generating unit 16 outputs the generated circuit symbol data to the information processing terminal 60.

[0036] The DRC parameter generating unit 17 generates DRC parameters in a predetermined format using the generated table sheet. The DRC parameter generating unit 17 generates circuit symbol data using data in the common part and the DRC parameter generation dedicated part of the table sheet. The DRC parameter generating unit 17 generates DRC parameters in accordance with a predetermined format of the CAD device. The DRC parameter generating unit 17 outputs the generated DRC parameters to the information processing terminal 60.

[0037] The storage unit 18 stores information necessary for processing of the information generating device 10. For example, the storage unit 18 stores a trained model, extracted information, a table sheet, and a program for converting the contents of the table sheet used by the extraction unit 12 into the format of a CAD device. The above-mentioned information stored in the storage unit 18 is an example, and the storage unit 18 may store other information in addition to the above-mentioned information.

[0038] (operation) Next, an example of the operation of the information generating device 10 according to the present embodiment will be described with reference to the drawings. Figures 10 and 11 are flowcharts showing an example of the operation of the information generating device according to the present disclosure.

[0039] 10, the analysis unit 11 acquires text information and layout information of a data sheet 70 (step S11). First, the analysis unit 11 acquires a data sheet from the data sheet storage device 110. If the content of the data sheet 70 is in a PDF format including text information, layout information, and the like, the analysis unit 11 acquires the text information and layout information incorporated in the PDF. If the content of the data sheet 70 is in a format consisting of only image information, the analysis unit 11 acquires the text information and layout information using known optical character recognition technology, natural language processing technology, and image processing technology.

[0040] Next, the extraction unit 12 extracts values ​​of items necessary for generating a table sheet from the data sheet 70 (step S12). The extraction unit 12 inputs the data sheet 70 and the text information, layout information, and table information identified by the analysis unit 11 to a trained model that has been machine-learned, and obtains values ​​of items necessary for generating a table sheet as output. The extraction unit 12 stores the extracted information in the storage unit 18 as extracted information. When extracting information, the trained model may also output a value indicating the reliability of the extracted information. The reliability is associated with the extracted information and stored in the storage unit 18.

[0041] Next, the input request unit 13 requests the information processing terminal 60 to input information related to the change of the extracted information. First, the input request unit 13 outputs the extracted information to the display unit of the information processing terminal 60 and requests an input related to the change of the extracted information (step S13). Next, the input request unit receives information related to the change of the extracted information transmitted from the information processing terminal 60 (step S14). The input request unit 13 changes the extracted information stored in the storage unit 18 using the received information.

[0042] Next, the input / output classification name acquiring unit 14 acquires the input classification name and the output classification name (step S15). The input / output classification name acquiring unit 14 searches the input / output classification sheet 80 using the values ​​of each item included in the extraction information stored in the storage unit 18. If there is an item on the input / output classification sheet that matches the extraction information, the input / output classification name acquiring unit 14 acquires the input classification name or output classification name corresponding to that item by linking it to the extraction information.

[0043] Next, the table sheet generating unit 15 generates a table sheet using the extraction information stored in the storage unit 18 and the input category name and output category name linked to the extraction information (step S16). The table sheet generating unit 15 generates the table sheet by combining the extraction information with the input category name and output category name.

[0044] Next, the circuit symbol generating unit 16 generates circuit symbol data using the generated table sheet (step S17).

[0045] Next, the DRC parameter generating unit 17 generates DRC parameters using the generated table sheet (step S18).

[0046] Next, the information generating device 10 outputs the generated information (step S19). Specifically, the information generating device 10 outputs at least one of the generated table sheet, the circuit symbol data, and the DRC parameters. After completing the process of step S18, the information generating device 10 ends the series of processes described above.

[0047] As described above, the information generating device of this embodiment includes an analysis unit, an extraction unit, an input / output category name acquisition unit, and a table sheet generating unit. The analysis unit acquires text information and layout information of a data sheet. The extraction unit extracts values ​​of items required for table sheet generation from the data sheet as extraction information. The input / output category name acquisition unit acquires input category names and output category names of each item included in the extraction information. The table sheet generating unit generates a table sheet using the extraction information and the input category names and output category names linked to the extraction information.

[0048] The information generating device of this embodiment can generate circuit symbol data and DRC parameters efficiently by the above configuration. This is because the generated table sheet includes information for generating circuit symbol data and information for generating DRC parameters. In general, when creating a library for electric circuit CAD, it is necessary to obtain pin information twice, when generating circuit symbol data and when generating DRC parameters, which is inefficient. In contrast, the information generating device of this embodiment only requires obtaining pin information when generating a table sheet, so there is no need to obtain pin information twice, and circuit symbol data and DRC parameters can be generated efficiently.

[0049] In addition, in the information generating device according to one aspect of the present embodiment, the extraction unit inputs the data sheet and the text information, layout information, and table information specified by the analysis unit into the machine-learned trained model, thereby obtaining values ​​of items necessary for generating a table sheet as an output. The trained model is trained using the search target, query, section name, and extraction item as training data. The trained model is also trained using the symbols included in the block diagram of the data sheet and the types of inputs and outputs corresponding to the symbols as training data. In this case, the trained model obtains the types of inputs and outputs of the pins related to the symbols as output, as extraction information. In general, in creating a library for electric circuit CAD, an operator searches the data sheet of a component using a query and inputs the values ​​of each item into the table sheet, which is inefficient. According to the information generating device according to the present embodiment, the trained model can obtain information to be input into the table sheet. As a result, according to the information generating device according to the present embodiment, the table sheet can be created efficiently without relying on human labor. Therefore, the information generating device according to the present embodiment can efficiently generate circuit symbol data and DRC parameters.

[0050] In addition, in the information generating device according to one aspect of the present embodiment, when extracting information, the trained model also outputs a value indicating the reliability of the extracted information. The reliability is a value that quantifies the likelihood of the value output by the trained model. This allows the likelihood of the value output by the trained model to be intuitively understood.

[0051] The information generating device according to one aspect of the present embodiment further includes an input request unit. The input request unit outputs the extraction information to the information processing terminal and requests the information processing terminal to input information related to the change of the extraction information. The information processing terminal that has received the input request complements the information that could not be extracted from the data sheet and corrects errors. The input request unit updates the extraction information stored in the storage unit using the received information related to the change. By including the input request unit, the information generating device according to the present embodiment can generate a table sheet using accurate information even if the value extracted by the trained model is incorrect.

[0052] In the information generating device according to one aspect of the present embodiment, the input request unit highlights the portion of the extracted information for which input is requested. For example, the input request unit highlights an item for which the extraction unit was unable to extract a value and an item with low reliability. This allows the operator of the information processing terminal that has received the input request to intuitively grasp the portion of the extracted information that should be changed.

[0053] In the information generating device according to the aspect of the present embodiment, the trained model is re-trained using the updated rule information as training data, thereby improving the determination accuracy of the trained model.

[0054] The information generating device of this embodiment can also efficiently generate circuit symbol data and DRC parameters for the following reasons. Generally, input of input category names and output category names is inefficient, since the input / output category sheet is searched manually and entered by a person. According to the information generating device of this embodiment, input category names and output category names corresponding to rule information are assigned, so that table sheets can be efficiently created. Therefore, the information generating device of this embodiment can efficiently generate circuit symbol data and DRC parameters.

[0055] The information generating device according to one aspect of the present embodiment further includes a circuit symbol generating unit and DRC parameters. The circuit symbol generating unit generates circuit symbol data in a predetermined format. The DRC parameter generating unit generates DRC parameters in a predetermined format. The information generating device according to the present embodiment can generate circuit symbol data and DRC parameters that match the predetermined format. As a result, an information processing terminal that receives the generated circuit symbol data and DRC parameters can read the circuit symbol data and DRC parameters into the information generating device without the need for special processing.

[0056] Second embodiment

[0057] (composition) Next, the configuration of the information generating device according to this embodiment will be described with reference to the drawings. The information generating device 20 has a simplified configuration of the information generating device according to the first embodiment. In the following description, the description of the same parts as those in the first embodiment will be omitted as appropriate.

[0058] FIG. 11 is a block diagram showing an example of the configuration of the information generating device according to the present disclosure. The information generating device 20 includes an analysis unit 21, an extraction unit 22, an input / output classification name acquisition unit 24, and a table sheet generation unit 25. The analysis unit 21 acquires text information of a data sheet and layout information that is information in which an area of ​​the data sheet is divided into sections. The extraction unit 22 inputs the data sheet, the text information, and the layout information into a machine-learned trained model, and obtains extraction information that is the value of an item required for generating a table sheet as an output. The trained model is trained using as teacher data a search target that is the item name of the table sheet, a query used when extracting the extraction information, a section name that is the name of a section that includes the search target, and an extraction item that is an item to be extracted as the value of the search target. The input / output classification name acquisition unit 24 acquires input classification names and output classification names corresponding to the items of the extraction information by linking them to the extraction information from an input / output classification sheet that is a sheet used to search for input classification names and output classification names that are items required for generating design rule check parameters, using the extraction information. The table sheet generating unit 25 generates a table sheet using the extracted information and the input category name and output category name linked to the extracted information.

[0059] (operation) Next, an example of the operation of the information generating device according to the present disclosure will be described with reference to FIG.

[0060] First, the analysis unit 21 acquires text information of a data sheet and layout information that is information dividing an area of ​​the data sheet into sections (step S21).

[0061] Next, the extraction unit 22 inputs the data sheet 70, text information, and layout information into the machine-learned trained model, and obtains, as an output, extracted information that is the value of an item required to generate a table sheet (step S22). The trained model is trained using as training data the search target that is the item name of the table sheet, the query used when extracting the extracted information, the section name that is the name of the section that includes the search target, and the extracted item that is the item to be extracted as the value of the search target.

[0062] Next, the input / output classification name acquisition unit 24 uses the extraction information to acquire input classification names and output classification names corresponding to the items of the extraction information by linking them to the extraction information from the input / output classification sheet, which is a sheet used to search for input classification names and output classification names, which are items necessary for generating design rule check parameters (step S23).

[0063] Next, the table sheet generating unit 25 generates a table sheet using the extracted information and the input category name and output category name linked to the extracted information (step S24).

[0064] The information generating device of this embodiment generates a table sheet using a data sheet and an input classification sheet. The table sheet includes information for generating circuit symbol data and information for generating design rule check (DRC) parameters. In a typical electric circuit design process, there is a problem that pin information needs to be acquired twice, when generating circuit symbol data and when generating DRC parameters, which is inefficient. In contrast, the information generating device of this embodiment eliminates the need to acquire pin information twice, making it possible to efficiently generate circuit symbol data and DRC parameters.

[0065] (Hardware configuration) The functions of each of the components according to each of the embodiments of the present disclosure described above can be realized not only by hardware but also by a computer device and firmware under program control.

[0066] Fig. 13 is a diagram showing an example of a hardware configuration in which an information generating device according to the present disclosure is realized by a computer device 90 including a processor. The information generating device of each embodiment is realized by the computer device 90. As shown in Fig. 13, the computer device 90 includes a CPU (Central Processing Unit) 91, a memory 92, a storage device 93 such as a hard disk for storing programs, an input / output interface 94 for connecting input devices and output devices, and a communication interface 95 for connecting to a network.

[0067] The CPU 91 operates an operating system to control the information generating device of the present disclosure. For example, the CPU 91 reads out programs and data from a storage medium attached to a drive device or the like into the memory 92. The CPU 91 also functions as a part of the analysis unit 11, extraction unit 12, input request unit 13, input / output classification name acquisition unit 14, table sheet generation unit 15, circuit symbol generation unit 16, DRC parameter generation unit 17, and storage unit 18 of the information generating device 10 of the present disclosure, and executes processes or commands based on the programs.

[0068] The storage device 93 is, for example, an optical disk, a flexible disk, a magneto-optical disk, an external hard disk, or a semiconductor memory. A part of the storage device is a storage medium that is a non-volatile storage device, and the program is recorded therein. The program may also be downloaded from an external computer (not shown) connected to a communication network.

[0069] The input device connected to the input / output interface 94 is realized by, for example, a mouse, a keyboard, etc., and is used for input operations. Similarly, the output device connected to the input / output interface 94 is realized by, for example, a display, etc., and is used for displaying and checking output results.

[0070] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-mentioned embodiments. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be appropriately combined with other embodiments.

[0071] Some or all of the above embodiments can be described as follows. (Appendix 1) an analysis means for acquiring text information of a data sheet for each component used in designing an electric circuit and layout information, which is information dividing an area of ​​the data sheet into sections; an extraction means for inputting the data sheet, the text information, and the layout information into a machine-learned model, and obtaining, as an output, extracted information that is the values ​​of items required for generating a table sheet that stores information used to generate circuit symbol data and design rule check parameters; an input / output classification name acquiring means for acquiring, by using the extracted information, input classification names and output classification names corresponding to the items of the extracted information from an input / output classification sheet which is a sheet used for searching for input classification names and output classification names which are items necessary for generating the design rule check parameters, by linking the input classification names and output classification names corresponding to the items of the extracted information with the extracted information; a table sheet generating means for generating the table sheet by using the extracted information and the input classification name and the output classification name associated with the extracted information; An information generating device comprising: (Appendix 2) The information generating device described in Appendix 1, wherein the trained model is trained using as training data a search target which is an item name of the table sheet, a query used when extracting the extracted information, a section name which is the name of the section in which the search target is included, and an extraction item which is an item to be extracted as a value of the search target. (Appendix 3) The information generating device described in Appendix 1 or 2, wherein the trained model is trained using symbols included in the block diagram of the data sheet and the types of input / output corresponding to the symbols as training data, and obtains as an output the types of input / output of pins related to the symbols as the extracted information. (Appendix 4) 4. The information generating device of claim 3, wherein the symbol is an arrow including a direction of the arrow, and the type of input / output corresponding to the symbol is the type of input / output corresponding to the direction of the arrow. (Appendix 5) a circuit symbol generating means for generating the circuit symbol data in a predetermined format using the table sheet; a design rule check parameter generating means for generating the design rule check parameters of the format using the table sheet; 5. The information generating device according to claim 1, further comprising: (Appendix 6) 6. The information generating device according to any one of claims 1 to 5, further comprising an input requesting means for outputting the extraction information to an information processing terminal, requesting the information processing terminal to input information related to changing the extraction information, and changing the extraction information using the received input information. (Appendix 7) The information generating device described in Appendix 6, wherein the trained model also outputs a reliability which is a value that quantifies the likelihood of the extracted information, and the extraction means links the extracted information and the reliability and stores them in a memory unit. (Appendix 8) The input request means includes: 8. The information generating device according to claim 7, further comprising: highlighting, among the extracted information, items for which the extracted information could not be extracted from the data sheet and items with low reliability. (Appendix 9) The information generating device according to any one of appendix 6 to 8, wherein the trained model is re-trained using the changed extracted information as training data. (Appendix 10) The computer acquiring text information of the data sheet and layout information, which is information dividing an area of ​​the data sheet into sections, from a data sheet for each component used in designing an electric circuit; The data sheet, the text information, and the layout information are input into a machine-learned trained model; As an output from the trained model, extracted information is obtained, which is the values ​​of items required for generating a table sheet that stores information used to generate circuit symbol data and design rule check parameters; obtaining, by using the extracted information, input classification names and output classification names corresponding to the items of the extracted information from an input / output classification sheet, which is a sheet used for searching input classification names and output classification names that are items necessary for generating the design rule check parameters, by linking the input classification names and output classification names to the extracted information; The table sheet is generated using the extracted information and the input classification name and the output classification name associated with the extracted information. Information generation method. (Appendix 11) The information generation method described in Appendix 10, wherein the trained model is trained using as training data a search target which is an item name of the table sheet, a query used when extracting the extracted information, a section name which is the name of the section in which the search target is included, and an extraction item which is an item to be extracted as a value of the search target. (Appendix 12) The information generating method described in Appendix 10 or 11, wherein the trained model is trained using symbols included in the block diagram of the data sheet and the types of input / output corresponding to the symbols as training data, and obtains as output the types of input / output of pins related to the symbols as the extracted information. (Appendix 13) 13. The information generating method of claim 12, wherein the symbol is an arrow including a direction of the arrow, and the type of input / output corresponding to the symbol is the type of input / output corresponding to the direction of the arrow. (Appendix 14) The computer further Using the table sheet, generate the circuit symbol data in a predetermined format; 14. The information generating method according to any one of claims 10 to 13, further comprising generating the design rule check parameters of the format using the table sheet. (Appendix 15) The computer further outputting the extracted information to an information processing terminal; requesting the information processing terminal to input information related to the change of the extracted information; 15. The information generating method according to any one of claims 10 to 14, further comprising modifying the extracted information using the received input information. (Appendix 16) The information generation method described in Appendix 15, wherein the trained model also outputs a reliability, which is a value that quantifies the likelihood of the extracted information, and the extraction means links the extracted information and the reliability and stores them in a memory unit. (Appendix 17) 17. The information generating method of claim 16, wherein in outputting the extracted information, the computer highlights items of the extracted information for which the extracted information could not be extracted from the data sheet and items with low reliability. (Appendix 18) 18. The information generating method according to any one of claims 15 to 17, wherein the trained model is re-trained using the changed extracted information as training data. (Appendix 19) On the computer, A process of acquiring text information of a data sheet for each component used in designing an electric circuit and layout information, which is information dividing an area of ​​the data sheet into sections; A process of inputting the data sheet, the text information, and the layout information into a machine-learned model; A process of obtaining extracted information, which is the values ​​of items required for generating a table sheet that stores information used to generate circuit symbol data and design rule check parameters, as an output from the trained model; a process of acquiring, by using the extracted information, input classification names and output classification names corresponding to the items of the extracted information from an input / output classification sheet, which is a sheet used for searching for input classification names and output classification names that are items necessary for generating the design rule check parameters, by linking the input classification names and output classification names corresponding to the items of the extracted information with the extracted information; and executing a process of generating the table sheet by using the extracted information and the input classification name and the output classification name linked to the extracted information. program. (Appendix 20) The program described in Appendix 19, wherein the trained model is trained using as training data a search target which is an item name of the table sheet, a query used when extracting the extracted information, a section name which is the name of the section in which the search target is included, and an extraction item which is an item to be extracted as a value of the search target. (Appendix 21) The program described in Appendix 19 or 20, wherein the trained model is trained using symbols included in the block diagram of the data sheet and the types of input / output corresponding to the symbols as training data, and obtains as output the types of input / output of pins related to the symbols as the extracted information. (Appendix 22) 22. The program of claim 21, wherein the symbol is an arrow including a direction of the arrow, and the type of input / output corresponding to the symbol is a type of input / output corresponding to the direction of the arrow. (Appendix 23) The computer, in addition, A process of generating the circuit symbol data in a predetermined format using the table sheet; 23. The program according to claim 19, further comprising: a process for generating the design rule check parameters of the format using the table sheet. (Appendix 24) The computer, in addition, outputting the extracted information to an information processing terminal; a process of requesting the information processing terminal to input information related to a change in the extracted information; and executing a process of changing the extraction information using the received input information. (Appendix 25) The program described in Appendix 24, wherein the trained model also outputs a reliability, which is a value that quantifies the likelihood of the extracted information, and the extraction means links the extracted information and the reliability and stores them in a memory unit. (Appendix 26) The program of claim 25, wherein in outputting the extracted information, the computer highlights, among the extracted information, items for which the extracted information could not be extracted from the data sheet and items with low reliability. (Appendix 27) 27. The program according to any one of appendices 24 to 26, wherein the trained model is re-trained using the changed extracted information as training data. [Explanation of symbols]

[0072] 1 Information Generation System 10, 20 Information generation device 11, 21 Analysis Department 12, 22 Extraction part 13 Input request section 14, 24 Input / output classification name acquisition part 15, 25 Table sheet generation section 16 Circuit symbol generator 17 DRC (Design Rule Check) parameter generation section 18 Memory section 60 Information processing terminal 70 Datasheets 80 Input / Output Classification Sheet 90 Computer Equipment 91 CPU(Central Processing Unit) 92 Memory 93 Storage device 94 Input / Output Interface 95 Communication Interface

Claims

1. an analysis means for acquiring text information of a data sheet for each component used in designing an electric circuit and layout information, which is information dividing an area of ​​the data sheet into sections; an extraction means for inputting the data sheet, the text information, and the layout information into a machine-learned model, and obtaining, as an output, extracted information that is the values ​​of items required for generating a table sheet that stores information used to generate circuit symbol data and design rule check parameters; an input / output classification name acquiring means for acquiring, by using the extracted information, input classification names and output classification names corresponding to the items of the extracted information from an input / output classification sheet which is a sheet used for searching for input classification names and output classification names which are items necessary for generating the design rule check parameters, by linking the input classification names and output classification names corresponding to the items of the extracted information with the extracted information; a table sheet generating means for generating the table sheet by using the extracted information and the input classification name and the output classification name associated with the extracted information; An information generating device comprising:

2. The information generating device of claim 1, wherein the trained model is trained using as training data a search target which is an item name of the table sheet, a query used when extracting the extracted information, a section name which is the name of the section in which the search target is included, and an extraction item which is an item to be extracted as a value of the search target.

3. 3. The information generating device according to claim 1 or 2, wherein the trained model is trained using symbols included in the block diagram of the data sheet and types of input / output corresponding to the symbols as training data, and obtains as output the types of input / output of pins related to the symbols as the extracted information.

4. The information generating device according to claim 3 , wherein the symbol is an arrow including a direction of the arrow, and the type of input / output corresponding to the symbol is a type of input / output corresponding to the direction of the arrow.

5. a circuit symbol generating means for generating the circuit symbol data in a predetermined format using the table sheet; a design rule check parameter generating means for generating the design rule check parameters of the format using the table sheet; The information generating device according to claim 1 , further comprising:

6. 2. The information generating device according to claim 1, further comprising an input requesting means for outputting the extraction information to an information processing terminal, requesting the information processing terminal to input information related to changing the extraction information, and changing the extraction information using the received input information.

7. 7. The information generating device according to claim 6, wherein the trained model also outputs a reliability which is a value that quantifies the likelihood of the extracted information, and the extraction means links the extracted information and the reliability and stores them in a memory unit.

8. The input request means includes: The information generating device according to claim 7 , wherein, of the extracted information, items for which the extracted information could not be extracted from the data sheet and items with low reliability are highlighted.

9. The computer acquiring text information of the data sheet and layout information, which is information dividing an area of ​​the data sheet into sections, from a data sheet for each component used in designing an electric circuit; The data sheet, the text information, and the layout information are input into a machine-learned trained model; As an output from the trained model, extracted information is obtained, which is the values ​​of items required for generating a table sheet that stores information used to generate circuit symbol data and design rule check parameters; using the extracted information, from an input / output classification sheet that is a sheet used for searching for input classification names and output classification names that are items necessary for generating the design rule check parameters, to obtain input classification names and output classification names corresponding to the items of the extracted information by linking them with the extracted information; The table sheet is generated using the extracted information and the input classification name and the output classification name associated with the extracted information. Information generation method.

10. On the computer, A process of acquiring text information of a data sheet for each component used in designing an electric circuit and layout information, which is information dividing an area of ​​the data sheet into sections; A process of inputting the data sheet, the text information, and the layout information into a machine-learned model; A process of obtaining extracted information, which is the values ​​of items required for generating a table sheet that stores information used to generate circuit symbol data and design rule check parameters, as an output from the trained model; a process of acquiring, by using the extracted information, input classification names and output classification names corresponding to the items of the extracted information from an input / output classification sheet, which is a sheet used for searching input classification names and output classification names, which are items necessary for generating the design rule check parameters, by linking the input classification names and output classification names corresponding to the items of the extracted information with the extracted information; and executing a process of generating the table sheet by using the extracted information and the input classification name and the output classification name linked to the extracted information. program.

Images