A karshen watan Mayu, daliban da suka sauke karatu daga Technopark (Bauman MSTU), Technotrack (MIPT), Technosphere (Lomonosov Moscow State University) da Technopolis (Peter the Great St. Petersburg Polytechnic University) sun kare ayyukan difloma. An ware watanni uku don yin aiki, kuma mutanen sun saka hannun jari a cikin ƙwararrun iliminsu da ƙwarewar da suka samu sama da shekaru biyu na karatu.

A cikin duka, akwai ayyuka 13 akan tsaro, magance matsaloli daban-daban a masana'antu daban-daban. Misali:

• ajiyar girgije tare da ɓoyayyen fayil ɗin ɓoye;
• dandamali don ƙirƙirar bidiyo masu hulɗa (tare da ƙarewa daban-daban);
• allon wayo don wasa da darasi na gaske akan hanyar sadarwa;
• gine-gine don maido da hankali na labaran likitanci;
• Software don koyar da yara 'yan makarantar firamare tushen algorithmization.

Kazalika ayyukan daga sassan kasuwanci:

• Tsarin CRM don manzo TamTam;
• sabis na yanar gizo don bincika hotuna masu jigo akan taswira don Odnoklassniki;
• adireshin geocoding sabis na MAPS.ME.

A yau za mu gaya muku dalla-dalla game da ayyuka guda biyar na waɗanda suka kammala karatunmu.

Bincike na hankali na labaran likitanci

Akwai fagage da dama a fannin kimiyya, a kowanne daga cikinsu ana gudanar da bincike, ana buga kasidu masu yawa a cikin mujallu iri-iri. Waɗannan su ne fasahar bayanai, kimiyyar lissafi, lissafi, ilmin halitta, likitanci da sauran su.

Authors aikin yanke shawarar mayar da hankali kan fannin likitanci. Kusan duk labarai akan batutuwan likita ana tattara su akan tashar PubMed. Portal tana ba da nata binciken. Duk da haka, ƙarfinsa yana da iyaka. Sabili da haka, mutanen sun inganta tsarin bincike, sun kara goyon baya ga dogon tambayoyi da kuma ikon tace tambayoyin ta amfani da samfurin magana.

SERP ta ƙunshi jerin jerin takardu tare da fayyace batutuwan su, kuma ana ba da fifikon kalmomi da kalmomin da suka danganci waɗannan batutuwa ta amfani da ƙirar jigo mai yiwuwa. Mai amfani zai iya danna kan fitattun sharuɗɗan don taƙaita tambayar nema.

Don yin bincike ta cikin babbar rumbun adana bayanan PubMed cikin sauri, marubutan sun rubuta injin binciken nasu wanda za'a iya haɗa shi cikin sauƙi cikin kowane kayan more rayuwa.

Ana gudanar da binciken ne a matakai uku:

1. Ana zaɓar takaddun ɗan takara ta amfani da juzu'i.
2. An jera 'yan takarar ta hanyar amfani da algorithm BM25F, wanda ke yin la'akari da fannoni daban-daban a cikin takardu yayin binciken. Don haka, kalmomi a cikin taken suna da nauyi fiye da kalmomin da ke cikin zayyana.
3. Hakanan ana amfani da tsarin caching don hanzarta aiwatar da buƙatu akai-akai.

Gine-gine na Microservice:

Ainihin, tsarin bayanan rubutu ana canjawa wuri tsakanin ayyuka. Don babban saurin canja wuri, ana amfani da GRPC - tsarin haɗa kayayyaki a cikin gine-ginen microservice. Hakanan ana amfani da serialization bayanai ta amfani da tsarin musayar saƙon Protobuf.

Wadanne abubuwa tsarin ya ƙunshi:

• Sabar don sarrafa buƙatun mai shigowa mai shigowa akan Node.js.
• Load da buƙatun daidaitawa ta amfani da sabar wakili na nginx.
• Sabar Flask tana aiwatar da REST API kuma tana karɓar buƙatun da aka tura daga Node.js.
• Duk danyen bayanan da aka sarrafa, da kuma bayanan tambaya, ana adana su a MongoDB.
• Duk buƙatun don sakamakon da ya dace don haɓaka daftarin aiki je zuwa RabbitMQ.

Misalin sakamakon bincike:

Abin da muke shirin yi na gaba:

• Shawarwari lokacin tattara bita akan wani batu da aka bayar (gano mahimman batutuwa a cikin takarda da bincike ta hanyar sassan takardu).
• Bincika fayilolin PDF.
• Bangaren rubutun Semantic.
• Bibiyar batutuwa da abubuwan da ke faruwa akan lokaci.

Ƙungiyar aikin: Fedor Petryaykin, Vladislav Dorozhinsky, Maxim Nakhodnov, Maxim Filin

Toshe Log

A yau, lokacin da ake koyar da shirye-shirye da kimiyyar kwamfuta, yaran da suka isa makarantar firamare (masu aji 5-7) suna fuskantar matsalar sarrafa kayan. Bugu da ƙari, idan ɗalibai suna son kammala ayyuka a gida, dole ne su sanya ƙarin software akan kwamfutocin su. Dole ne malamai su bincika ɗimbin nau'ikan hanyoyin magance matsaloli iri ɗaya, kuma a yanayin koyo daga nesa, suma dole ne su samar da wata hanya don karɓar ayyuka daga ɗalibai.

Mawallafa na Block Log aikin sun zo ga ƙarshe: lokacin da suke koyar da yara na shekarun makarantar firamare abubuwan da ake amfani da su na algorithmization, bai kamata a mayar da hankali ga haddace umarnin harshe na shirye-shirye ba, amma akan gina zane-zane na algorithm. Wannan zai ba wa ɗalibai damar yin amfani da lokaci da ƙoƙari kan ƙirƙira algorithm, maimakon buga a cikin tsarukan daidaitawa.

Platform Toshe Log damar:

1. Ƙirƙiri kuma gyara taswirar tafiya.
2. Gudun abubuwan da aka ƙirƙira don ganin sakamakon aikin su (bayanan fitarwa).
3. Ajiye ku loda ayyukan da aka ƙirƙira.
4. Zana hotunan raster (ƙirƙirar hoto dangane da algorithm wanda yaron ya ƙirƙira).
5. Karɓi bayanai game da rikitarwa na algorithm da aka ƙirƙira (bisa yawan ayyukan da aka yi a cikin algorithm).

Ana sa ran rarraba ayyuka zuwa malamai da ɗalibai. Duk wani mai amfani yana karɓar matsayin ɗalibi; don samun matsayin malami, dole ne ka tuntuɓi mai gudanar da tsarin. Malami ba zai iya shigar da kwatancin da yanayin matsalolin kawai ba, amma kuma ya ƙirƙiri gwaje-gwaje na atomatik wanda za a ƙaddamar da shi ta atomatik lokacin da ɗalibi ya ba da mafita ga matsalar a cikin tsarin.

Editan Block Log Edita:

Bayan warware matsalar, ɗalibin zai iya saukar da maganin kuma ya ga sakamakon:

Dandalin ya ƙunshi aikace-aikacen gaba-gaba a cikin Vue.js da aikace-aikacen ƙarshen ƙarshen a cikin Ruby akan Rails. Ana amfani da PostgreSQL azaman bayanan bayanai. Don sauƙaƙe turawa, duk abubuwan da aka gyara tsarin ana tattara su a cikin kwantena Docker kuma an haɗa su ta amfani da Docker Compose. Sigar tebur na Block Log ya dogara ne akan tsarin Electron. An yi amfani da fakitin yanar gizo don gina lambar JavaScript.

Ƙungiyar aikin: Alexander Barulev, Maxim Kolotovkin, Kirill Kucherov.

Tsarin CRM don TamTam messenger

CRM kayan aiki ne don dacewa da hulɗa tsakanin kasuwanci da masu amfani da TamTam. An aiwatar da ayyuka masu zuwa:

• Maginin bot wanda ke ba ku damar ƙirƙirar bots ba tare da ƙwarewar shirye-shirye ba. A cikin 'yan mintoci kaɗan za ku iya samun cikakken bot wanda ba zai iya nuna wasu bayanai kawai ga masu amfani ba, har ma tattara bayanai, ciki har da. fayilolin da mai gudanarwa zai iya dubawa daga baya.
• RSS. Kuna iya haɗa RSS cikin sauƙi zuwa kowace tashoshi.
• An jinkirta aikawa. Yana ba ku damar aikawa da share saƙonni a lokutan da aka saita.

Har ila yau, ƙungiyar ta shiga cikin gwajin Bot API, ƙirƙirar bots da yawa da aka rubuta, kamar bot don gasar cin kofin duniya ta Hockey na 2019, bot don rajista / ba da izini a cikin sabis ɗinmu, da bot don CI/CD.

Kayan aikin mafita:

• Sabar gudanarwa ta ƙunshi tsarin sa ido ga kowane uwar garken da kowane akwati Docker akanta don gano matsala cikin sauri da dacewa da warware ta, duba ma'auni daban-daban da ƙididdigar amfani. Har ila yau, akwai tsarin gudanar da tsarin sarrafa aikace-aikacen mu mai nisa.
• Uwar garken tana ƙunshe da nau'in aikace-aikacenmu na yanzu, wanda ƙungiyar haɓaka ke samuwa don gwaji na gaba ɗaya.
• Ana samun sabar gudanarwa da tsarawa kawai ta hanyar VPN ga masu haɓakawa, kuma uwar garken samarwa ya ƙunshi sigar sakin aikace-aikacen. An keɓe shi daga hannun masu haɓakawa kuma yana samuwa ga mai amfani kawai.
• An aiwatar da tsarin CI/CD ta amfani da Github da Travis, sanarwa ta amfani da bot na al'ada a TamTam.

Tsarin gine-ginen aikace-aikacen tsari ne na zamani. Ana ƙaddamar da aikace-aikacen, bayanan bayanai, manajan daidaitawa da saka idanu a cikin kwantena Docker daban-daban, wanda ke ba ku damar zayyanawa daga yanayin ƙaddamarwa, canza ko sake kunna akwati daban. Ƙirƙirar topology na cibiyar sadarwa da sarrafa kwantena ana yin su ta amfani da Docker Compose.

Ƙungiyar aikin: Alexey Antufiev, Egor Gorbatov, Alexey Kotelevsky.

ForkMe

Aikin ForkMe dandamali ne na kallon bidiyo mai ma'amala, inda zaku iya ƙirƙirar bidiyon ku kuma ku nuna shi ga abokanku. Me yasa muke buƙatar bidiyo mai hulɗa idan akwai na yau da kullun?

Matsalolin da ba na layi ba na bidiyo da kuma damar da za a zabi ci gaba da kansu suna ba da damar mai kallo ya shiga ciki, kuma masu kirkiro abun ciki za su iya nuna labarun musamman, wanda masu amfani za su rinjayi makircinsu. Har ila yau, masu ƙirƙira abun ciki, ta hanyar nazarin kididdigar juyar da bidiyo, za su iya fahimtar abin da ya fi sha'awar masu sauraro kuma su sa kayan su zama masu ban sha'awa.

Lokacin haɓaka aikin, mutanen sun sami wahayi ta hanyar fim ɗin hulɗar Bandersnatch daga Netflix, wanda ya sami ra'ayoyi da yawa da kuma sake dubawa mai kyau. Lokacin da aka riga aka rubuta MVP, labarai sun bayyana cewa Youtube na shirin ƙaddamar da wani dandamali don jerin shirye-shirye, wanda ya sake tabbatar da shaharar wannan jagorar.

MVP ya haɗa da: mai kunna ma'amala, mai gina bidiyo, bincike ta abun ciki da tags, tarin bidiyo, sharhi, ra'ayoyi, ƙima, tashoshi da bayanan bayanan mai amfani.

Tarin fasaha da aka yi amfani da shi a cikin aikin:

Yaya aka tsara don haɓaka aikin:

• tarin ƙididdiga da bayanai game da sauyawa zuwa bidiyo;
• sanarwa da saƙonnin sirri don masu amfani da rukunin yanar gizon;
• versions ga Android da iOS.

Bayan wannan muna shirin ƙara:

• ƙirƙirar labarun bidiyo daga wayarka;
• gyara gutsure na bidiyo da aka sauke (misali trimming);
• ƙirƙira da ƙaddamar da tallace-tallacen hulɗa a cikin mai kunnawa.

Ƙungiyar aikin: Maxim Morev (cikakken mai haɓakawa, ya yi aiki a kan gine-ginen aikin) da Roman Maslov (cikakken mai haɓakawa, ya yi aiki a kan tsarin aikin).

Kan-Layi-On-Board

A yau, iyaye suna mai da hankali sosai ga ci gaban tunanin 'ya'yansu, kuma yara suna sha'awar wasanni na hankali. Saboda haka, chess yana sake samun farin jini. Kuma ko da yake dara a gabaɗaya ya shahara sosai, samun abokin gaba na yau da kullun don wasanni yana da matsala. Saboda haka, mutane da yawa suna amfani da sabis na dara na kan layi, duk da cewa yawancin 'yan wasa sun fi son yin wasa "rayuwa" tare da ainihin guda. Duk da haka, yayin wasa da dara, mutum yana yin ƙoƙari sosai a hankali kuma ya gaji, kuma wannan gajiyar yana cike da mummunan tasirin zama a kwamfuta ko smartphone. Sakamakon haka, kwakwalwa takan yi nauyi bayan wasanni biyu kacal.

Duk waɗannan abubuwan sun tura marubutan zuwa ra'ayin aikin Kan-Layi-On-Board, wanda ya ƙunshi sassa uku: allon chess na zahiri, aikace-aikacen tebur da sabis na yanar gizo. Jirgin filin wasa ne na yau da kullun, wanda ke gane matsayin yanki kuma, tare da taimakon alamar haske, yana nuna motsin abokin hamayya. Ana haɗa allon ta USB zuwa PC kuma yana sadarwa tare da aikace-aikacen tebur. A cikin yanayin horarwa (da kuma na yara), ana ba da haske ga yuwuwar motsinku.

Aikace-aikacen yana ɗaukar mahimman ayyuka na gudanar da hukumar, wanda ke ba ku damar rage farashinsa sosai da kawo aiwatar da mafi yawan ayyuka zuwa matakin software. Aikace-aikacen yana sadarwa tare da sabis na yanar gizo wanda babban ƙimarsa shine sabuntawa mai ƙarfi.

Babban yanayin don amfani da samfurin: mutum ɗaya yana wasa akan sabis, na biyu akan allon jiki wanda aka haɗa da sabis ɗin. Wato, sabis ɗin yana ɗaukar aikin sadarwa.

Ƙungiyar aikin: Daniil Tuchin, Anton Dmitriev, Sasha Kuznetsov.

Kuna iya karanta ƙarin game da ayyukanmu na ilimi a wannan haɗin. Kuma a yawaita ziyartar tashar Technostream, sabbin bidiyoyi na ilmantarwa game da shirye-shirye, haɓakawa da sauran fannoni suna bayyana a can akai-akai.

source: www.habr.com