Federal gosudarstvennoe uchebnoe predpriyatie Chair of the System of Artificial Intelligence

Interaction with Neurogenesis
4. Aktivizirovat process modelirovaniya c pomoshchyu component
systems;
uproshcheniya programmirovaniya can be used ActiveX- components
testing models on testovyx primerax c pomoshchyu component
osushchestvlyaetsya by exchanging XML-soobshcheniyami on TCP / IP protocol. For
5. ServiceName - is used to name the service operatsionnoy
5. Dojdatsya zaversheniya protsessa modelirovaniya i vÿpolnit
maksimalnoe vremya ojidaniya otveta ot neyroservera v millisekundax
soobshcheniy. The following sections describe the components and functions.
Typical variants of neuroserve use:
In sluchae, esli na kompetutere ustanovleno neskolko protsessorov, to
1. Install and check the parameters of the connection with the neuroservice with
the help of the Connection component;
Connection
6. DisplayName - otobrajaemoe naimenovanie servisa operatsionnoy
system.
neurosurgeon client, hiding details of the set protocol and formats
Model.
vÿpolnyaetsya obrabotchik zaprosov (Host), port obrabotchika zaprosov (Port) i
ukazÿvaya razlichnÿe portÿ i nastroyki servisa operatsionnoy sistemy.
3. Add primers for classification with the help of Samples component;
The component serves for the organization of connections with the neuroservice.
Ispolzuetsya ostalnymi components. Method Login vÿpolnyaet ustanovku parameters
soedineniya s
2. Create a classification system with the help of the Samples component;
117
Machine Translated by Google

---

priznak klassifikatsii neizvesten, to sootvetstvuyushchaya position array
Check vÿpolnyaet check korrektnosti parameters peredannyx v
parameter, opredelyayushchiy sposob ukazaniya vesa primera (WhoWeighted). Esli
positive sign. On umolchaniyu vesa vsex primerov ravny 1.
doljen bit unikalnym. Klassÿ numeruyutsya posledovatelno from 0 to
GetServerVersion vozvrashchaet stroku s opisaniem versii neyroservera.
The method receives the code of the classification system (SysId) and the class number
GetSamples. The method at the entrance adopts the code of the classification system (SysId),
classification system codes (SysId), prime class number (ClassNum), prime label (Mark),
weight prime (Weight) and array of symbols
Component serves for registration, assignment of classifications and storage
The DelSys method removes the classification system. At this udalyayutsya everything
teacher. The method NewSys registers a new system of classification. Na
The method SetClassWeight establishes all the instructions specified
line in ANSI encoding, containing up to 100 characters. The tag may be a bit
class (ClassNum) and all prime class (ClassWeight). Weight is imet
mojno peredavat pustÿe stroki. Method nichego ne vozvrashchaet.
classes (ClassCount), the number of criteria for classification (PropCount) and
ActiveX, the line of perekodiruetsya tags in UNICODE format. Esli nekotoryy
Login method. Result proverki vozvrashchaetsya in vide logicheskoy velichiny.
WhoWeighted raven 0, to ves primerov ukazÿvaetsya na urovne klassa, esli 1, to ves
ukazÿvaetsya kajdomu primeru otdelno. Classification system code
GetClassWeight vozvrashchaet ves primerov ukazannogo klassa. Na vxode
Values must be set to NULL. Get early added
primers can be helped by the method
(ClassNum).
Primers are added by the method AddSample. Na vxode metod poluchaet
Samples
ClassCount -1.
primerov.
The Login method establishes the parameters of the connection with the neuroservice.
The input method receives a component of the Connection class. Neyroserver stroit modeli v
vide neyrosetevyx klassifikatorov s
svyazannÿe s sistemoy primery, pravila predobrabotki primerov i modeli. The input method
adopts the code udalyaemoy classification system (SysId).
classification (Values). Label primera predstavlyaet soboy proizvolnuyu
vxode metod prinimaet kod sistemy klassifikatsii (SysId), kolichestvo
dostupa ne vypolneno. In this regard, the quality of the name of the user and the password
class. The input method receives the classification system code (SysId), number
not unique. At the transfer of tags in the method, on the need for technology
118
Machine Translated by Google

---

Mojno takje poluchit tolko massiv metok s pomoshchyu metoda
Model
Get information about the state of the process of modeling mojno s
ispolzovaniya sozdannyx models. Models pozvolyayut on a set of signs
(SysId). On the output poluchaetsya massive method of primerov (Marks).
classification (SysId). Na vyxode method vozvrashchaet stroku, harakterizuyushchuyu
sostoyanie process modelirovaniya: "ANALYSIS" - vÿpolnyaetsya analysis task zadichi
modelirovaniya; "SYNTHESIS" - vÿpolnyaetsya
zavershen; “NONE” - modeling has not been completed.
method peredayutsya code system classification (SysId) and label primerov. Udalyayutsya
all primery, svyazannÿe s sistemoy klassifikatsii i metkoy.
FindModel. The input method receives the classification system code (SysId). Esli model
uje sushchestvuet, ee mojno polzovatsya vo vremya protsessa
GetExceptionClassName. GetErrorMessage vozvrashchaet soobshchenie ob oshibke, a
the model will be replaced by a new model. Esli na moment vyzova metoda uje
results can be used with the method StopFindModel. Na vxode metod
posted in ochred on vÿpolnenie.
nomerov klassov primerov (Y) i massiv metok primerov (Mark).
otsutstviya oshibok metodÿ vozvrashchayut pustÿe stroki.
promejutochnyx rezultatov mojno s pomoshchyu metoda PauseFindModel. Renew the
modeling process with the ContinueFindModel method.
GetMarks. At the entrance to the method neobxodimo transfer code classification system
The component serves for the management of the modeling process i
pomoshchyu method GetModelingState. The input method uses the system code
synthesis neyrosetevyx models; “COMPLETE” - modeling process
You can remove the primer with the help of DelSample method. Na vxode
predskazÿvat class of objects modeling. The Login
method establishes the parameters of the connection with the neuroservice. The
input method receives a component of the Connection class. Activation process
modelirovaniya osushchestvlyaetsya vyzovom
Dlya kontrolya oshibok sushchestvuyut dva metoda GetErrorMessage i
modeling. Po okonchaniyu process modelirovaniya sushchestvuyushchaya
Stop the process of modeling with poterey vsex promejutochnyx
GetExceptionClassName naimenovanie klassa isklyucheniya. In sluchae
a na vyxode array arrays znacheniy priznakov klassifikatsii (X), array
vÿpolnyaetsya drugoy process modelirovaniya, to zadacha modelirovaniya budet
receives the code of the classification system (SysId).
Priostanovit all processes of modeling with soxraneniem
119
Machine Translated by Google

---

correctness of parameters with the help of the method CheckComputer. Na vxode metod
After zaversheniya modelirovaniya mojno predskazÿvat class
otsutstviya oshibok metodÿ vozvrashchayut pustÿe stroki.
zadaniy ”.
neobxodimo use GetClassNum method. Na vxode metod prinimaet
Component serves to manage a list of computers
installed on it module "Executor of the task". On the input method
"Planner task". You can also use the regcomp utility (see installation instructions).
Sweden ob obpolzuemyx kompyuterax
classification (Values). Esli nekotoryy priznak klassifikatsii neizvesten, to
sootvetstvuyushchaya pozitsiya massiva Values doljna bÿt ustanovlena v
vychisleniy pri sozdanii neyrosetevyx modeley. Na kajdom ispolzuemom
Dlya kontrolya oshibok sushchestvuyut dva metoda GetErrorMessage i
zadaniy ”. Rekomenduetsya on one computer ustanavlivat stolko modules
resursy, neobxodimo posledovatelno vÿzvat metody PauseFindModel i
GetCount returns the number of modules used ”Ispolnitel
modeling.
GetExceptionClassName naimenovanie klassa isklyucheniya. In sluchae
Pered vypolneniem metoda AddComputer mojno proverit
modeliruemyx objects on the set of classification classifications. For this