[{"_1":2,"_412":-5,"_413":-5},"loaderData",{"_3":-5,"_4":-5,"_5":6},"routes/i18nLayout","routes/appGuard","routes/sharingView",{"_7":8,"_9":10},"kind","presentation","data",{"_11":12,"_13":14,"_15":-5,"_16":17,"_18":19,"_411":-5},"needPin",false,"chatId","5e17e82d-2a95-49c3-b876-ab2246e3163a","presentationId","title","ИИ и машинное обучение","slides",[20,58,82,93,138,152,174,201,220,238,259,287,308,338,354,372,391],{"_21":22,"_27":28,"_29":30,"_31":32,"_33":34,"_35":12,"_36":37,"_38":39,"_42":43,"_44":45,"_46":47,"_48":49,"_50":51,"_52":53,"_54":24,"_55":12,"_56":57},"infographicAssessment",{"_23":24,"_25":26},"priority","none","reason","","slide",1,"outlineId","fc5ee57a-0e0f-4678-9d33-0e0d7b1e6186","brief","# ИИ и машинное обучение\nСингин Игорь Александрович\nРуководитель группы","slideType","cover","isInfographic","infographicType","structure","slideLabel",{"_16":17,"_40":41},"shortSummary","Титульный слайд","imageRefs",[],"chartRefs",[],"tableRefs",[],"visualRefs",[],"chunkRefs",[],"previewUrl","https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/1/master/f6944072-b7ec-431a-b609-e5ec43979b96.webp","gate","isLocked","renderStatus","ready",{"_21":59,"_27":60,"_29":61,"_31":62,"_33":63,"_35":12,"_36":37,"_38":64,"_42":67,"_44":77,"_46":78,"_48":79,"_50":80,"_52":81,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},2,"f9f64416-8a76-4bb7-9f2b-8d18b64f2171","# Сингин Игорь Александрович\nРуководитель группы Продвинутая аналитика\n## Сингин Игорь Александрович\nРуководитель группы Продвинутая аналитика\nОбо мне\n\nВ ИТ с 2009 года\nМатематик, занимаюсь машинным обучение с 2015 года\nРаботал в на позициях Senior, Team Lead в банках, ИТ-компаниях\nЛидер компетенций Сибура по направлению «ИИ и машинное обучение»","team",{"_16":65,"_40":66},"Сингин Игорь Александрович\rРуководитель группы Продвинутая аналитика","Об авторе",[68],{"_69":70,"_71":72,"_73":74,"_75":76},"source","existing","imageId","f0e6e55f-5808-40c3-bc3b-3f8b1fa9f6be","imageType","portrait","description","portrait: A man wearing a blue sweater and striped shirt looking directly at the camera.",[],[],[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/2/master/b5158458-dae5-4ca8-9799-ba959ceb370d.webp",{"_21":83,"_27":84,"_29":85,"_31":62,"_33":63,"_35":12,"_36":37,"_38":86,"_42":87,"_44":88,"_46":89,"_48":90,"_50":91,"_52":92,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},3,"39c9aff6-c456-411b-b35c-4b656ba5bde8",{"_16":65,"_40":66},[],[],[],[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/3/master/265a6fb7-0e48-4ad7-872d-da29247796af.webp",{"_21":94,"_27":95,"_29":96,"_31":97,"_33":98,"_35":12,"_36":37,"_38":99,"_42":102,"_44":107,"_46":108,"_48":109,"_50":136,"_52":137,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},4,"32cd4de1-f47e-41ae-8f08-4bc9ad3fcba2","# О данной лекции\n4","agenda",{"_16":100,"_40":101},"О данной лекции","Введение",[103],{"_69":70,"_71":104,"_73":105,"_75":106},"50c99cdc-3822-45e4-a36b-0346b4e084ca","infographic","infographic: A humorous comparison of two types of machine learning articles: overly complex math vs exaggerated hype.",[],[],[110],{"_111":112,"_7":113,"_25":114,"_115":116,"_117":95,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":134},"refId","V4_0_45802f2d","smartart","unsupported_smartart","sourcePptxKey","storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/uploads/9a205f09-447f-4dc1-b64e-4d0471c0806f.pptx","slideNumber","shapeIndex",0,"shapePath","0","sourceSlideNumber","sourceShapeIndex","sourceShapePath","shapeName","Схема 13","className","SmartArt","contentType","source_shape","shapeWidth","shapeHeight","extractSourceData",{"_135":116,"_117":95,"_118":119,"_120":121,"_129":130},"pptxKey",[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/4/master/b55dc8f7-3076-4a36-b8c9-37925098e5fe.webp",{"_21":139,"_27":140,"_29":141,"_31":142,"_33":143,"_35":12,"_36":37,"_38":144,"_42":146,"_44":147,"_46":148,"_48":149,"_50":150,"_52":151,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},5,"e9c0f0ec-0681-420e-b2fc-a1e449171dc0","# О данной лекции\nО чём будет рассказ\n\nЧто такое ИИ, машинное обучение, глубокое обучение и как они соотносятся\nЗачем нужен ИИ, как именно ИИ приносит пользу\nКакие модели ИИ бывают, в чем их преимущества и недостатки\nЧто нужно для создания ИИ моделей\nКак ВП управлять созданием моделей и избегать ошибок\nКогда ИИ захватит мир\nО чём НЕ будет рассказано\n\nКак самому обучать ML модели\nПро математические аспекты машинного обучения\nВыведение моделей ИИ в production \nСложные формулы, интегралы, якобианы, доказательства теорем\nРассказа о всех существующих методах машинного обучения\n5","explanation",{"_16":100,"_40":145},"Содержание",[],[],[],[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/5/master/be41dcb7-1cee-4592-aade-18f703c486b0.webp",{"_21":153,"_27":154,"_29":155,"_31":156,"_33":143,"_35":157,"_36":37,"_38":158,"_42":161,"_44":166,"_46":167,"_48":168,"_50":172,"_52":173,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},6,"e15934d5-bda9-418d-8549-08b5eb71289b","# Что такое ИИ и машинное обучение\nИскусственный интеллект —свойство искусственных систем выполнять интеллектуальные функции, ранее присущие человеку. Это название всей области знаний, как биология или химия. Четкой формулировки данного термина не существует.\n### Нейронная сеть – это математическая модель, которая имитирует работу биологического мозга.\nМашинное обучение — это наука, изучающая алгоритмы, автоматически улучшающиеся благодаря опыту. Это одна из ветвей математики и науки об алгоритмах\n6",true,{"_16":159,"_40":160},"Что такое ИИ и машинное обучение","Определения",[162],{"_69":70,"_71":163,"_73":164,"_75":165},"85f36b66-2bc4-4df0-83fb-154751f127a8","diagram","diagram: Nested set diagram illustrating the hierarchy of Artificial Intelligence, Machine Learning, Neural Networks, and Deep Learning.",[],[],[169],{"_111":170,"_7":113,"_25":114,"_115":116,"_117":154,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":171},"V6_0_45802f2d",{"_135":116,"_117":154,"_118":119,"_120":121,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/6/master/8ec7ef8a-601d-45e9-9393-61bd206eeaf9.webp",{"_21":175,"_27":176,"_29":177,"_31":178,"_33":179,"_35":157,"_36":179,"_38":180,"_42":183,"_44":184,"_46":185,"_48":186,"_50":199,"_52":200,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},7,"12c44603-501b-4151-8ac2-ce10ff04b2b8","# Машинное обучение VS Классические алгоритмы\n### Классические алгоритмы\n### Данные\nАлгоритм\nРезультат ?\n### Машинное обучение\nДанные\nАлгоритм ?\nРезультат\n7","comparison",{"_16":181,"_40":182},"Машинное обучение VS Классические алгоритмы","Сравнение",[],[],[],[187,190],{"_111":188,"_7":113,"_25":114,"_115":116,"_117":176,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":189},"V7_0_45802f2d",{"_135":116,"_117":176,"_118":119,"_120":121,"_129":130},{"_111":191,"_7":192,"_25":193,"_115":116,"_117":176,"_118":194,"_120":195,"_122":-5,"_123":-5,"_124":26,"_125":196,"_127":197,"_129":130,"_131":119,"_132":119,"_133":198},"V7_15_45802f2d","ole","unsupported_ole_object",15,"15","Объект 5","OleObjectFrame",{"_135":116,"_117":176,"_118":194,"_120":195,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/7/master/5f2baf24-c213-48a5-9d79-cd8166c3b605.webp",{"_21":202,"_27":203,"_29":204,"_31":205,"_33":179,"_35":157,"_36":179,"_38":206,"_42":207,"_44":208,"_46":209,"_48":210,"_50":218,"_52":219,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},8,"9baa8de7-e122-4211-b92b-660e7f0aa707","# Машинное обучение VS Классические алгоритмы\n### Классические алгоритмы\n### Данные\nАлгоритм\nРезультат ?\n### Машинное обучение\nДанные\nАлгоритм ?\nРезультат\n8",{"_16":181,"_40":182},[],[],[],[211,214],{"_111":212,"_7":113,"_25":114,"_115":116,"_117":203,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":213},"V8_0_45802f2d",{"_135":116,"_117":203,"_118":119,"_120":121,"_129":130},{"_111":215,"_7":192,"_25":193,"_115":116,"_117":203,"_118":140,"_120":216,"_122":-5,"_123":-5,"_124":26,"_125":196,"_127":197,"_129":130,"_131":119,"_132":119,"_133":217},"V8_5_45802f2d","5",{"_135":116,"_117":203,"_118":140,"_120":216,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/8/master/55a08c60-fad9-418a-97f5-5957b8990652.webp",{"_21":221,"_27":222,"_29":223,"_31":224,"_33":143,"_35":12,"_36":37,"_38":225,"_42":228,"_44":233,"_46":234,"_48":235,"_50":236,"_52":237,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},9,"553e6ff3-5c16-4c28-a333-79b70cde994d","# Что нужно для обучения моделей\nНеобходимо\n\nЗадача с математической формулировкой\nДанные\nПризнаки\nВычислительные мощности – CPU, RAM, GPU.\nData Scientists (аналитики данных)\nВремя\n9",{"_16":226,"_40":227},"Что нужно для обучения моделей","Требования",[229],{"_69":70,"_71":230,"_73":231,"_75":232},"e738fa2b-c631-4b2c-9fd5-758a3cd5991b","photo","photo: Image: slide9_img3.png",[],[],[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/9/master/d2b27d02-681f-43ef-bfee-0a9fe3066677.webp",{"_21":239,"_27":240,"_29":241,"_31":242,"_33":243,"_35":157,"_36":37,"_38":244,"_42":247,"_44":254,"_46":255,"_48":256,"_50":257,"_52":258,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},10,"44dc004b-0f7a-48a8-a3af-262e380a797c","# Данные для машинного обучения-1\n10","gallery",{"_16":245,"_40":246},"Данные для машинного обучения-1","Данные",[248,251],{"_69":70,"_71":249,"_73":164,"_75":250},"e1cd2739-e41d-4049-a48d-d1cf4ed18780","diagram: Diagram illustrating the 'Garbage In-garbage Out' paradigm with a perfect model.",{"_69":70,"_71":252,"_73":105,"_75":253},"42593ef3-9132-4a2b-adad-9bc520de4716","infographic: Caricature illustrating the concept 'Garbage in, garbage out' with a person buried in trash at a computer.",[],[],[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/10/master/a960e2d7-b9b1-4082-a30f-da592be89d12.webp",{"_21":260,"_27":261,"_29":262,"_31":263,"_33":264,"_35":12,"_36":37,"_38":265,"_42":268,"_44":273,"_46":274,"_48":284,"_50":285,"_52":286,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},11,"2e6378ae-013e-4c3e-a14c-f918f5c42592","# Данные для машинного обучения-2\n11","pricing",{"_16":266,"_40":267},"Данные для машинного обучения-2","Объемы данных",[269],{"_69":70,"_71":270,"_73":271,"_75":272},"571c53fc-46d5-42df-b33f-bdc8e33afbdb","chart","chart: Donut charts showing three different splits of training, validation, and test data percentages.",[],[275],{"_276":277,"_278":279,"_280":281,"_282":283,"_115":116,"_117":261,"_118":84},"tableId","54f69398-60f8-4af8-bfcf-2bdb031fcb3a","tableHtml","

	Количество наблюдений	Метод	Результат
1	< 100		построение модели невозможно
2	100-1 000	линейные модели, дерево решений	слабая прогностическая способность
3	1000-10 000	бустинги, леса решений	средняя\rпрогностическая способность
4	10 000 – 1 млн.	бустинги, нейросети	хорошая прогностическая способность

","summary","| Количество наблюдений, Метод, Результат","requestId","33966e2b-f429-48b7-ab12-e1b41d285e0a",[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/11/master/a9ccae29-8f5e-480c-b5ea-ab4f701efdf6.webp",{"_21":288,"_27":289,"_29":290,"_31":291,"_33":292,"_35":12,"_36":37,"_38":293,"_42":296,"_44":300,"_46":301,"_48":302,"_50":306,"_52":307,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},12,"07256bda-f981-4e27-9930-a73238f747c6","# Данные и обучение моделей\n12","case_study",{"_16":294,"_40":295},"Данные и обучение моделей","Обучение",[297],{"_69":70,"_71":298,"_73":231,"_75":299},"1dc7adb9-8edd-4c9f-bb14-65dec7fbed7c","photo: Image: slide12_img8.png",[],[],[303],{"_111":304,"_7":113,"_25":114,"_115":116,"_117":289,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":305},"V12_0_45802f2d",{"_135":116,"_117":289,"_118":119,"_120":121,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/12/master/f71573e8-1fa7-44a4-ba11-8c80a190bef9.webp",{"_21":309,"_27":310,"_29":311,"_31":312,"_33":313,"_35":12,"_36":37,"_38":314,"_42":317,"_44":330,"_46":331,"_48":332,"_50":336,"_52":337,"_54":24,"_55":12,"_56":57},{"_23":24,"_25":26},13,"81ba6219-f4cf-4579-8fd1-835bde9e74d9","# Линейная регрессия\n### Преимущества\n\nМожет обучаться на малом наборе данных\nУстойчива к переобучению\nБыстрое обучение и применение\nМожет экстраполировать тренд\n### - факторы модели\n\\- параметры регрессии\nНедостатки\n\nНе может выявлять нелинейные взаимосвязи\nОбычно уступает по точности прогнозирования другим методам\n13","solution",{"_16":315,"_40":316},"Линейная регрессия","Модель",[318,321,324,327],{"_69":70,"_71":319,"_73":231,"_75":320},"19e77475-7e5f-4006-a563-f0115e5e68e5","photo: Image: slide13_img9.png",{"_69":70,"_71":322,"_73":271,"_75":323},"f811590b-9b0f-4079-86c2-c8def4cb6bd7","chart: Scatter plots comparing linear and polynomial regression for traffic predictions.",{"_69":70,"_71":325,"_73":231,"_75":326},"1aaa1783-f0f7-41ef-ac54-639f6703531d","photo: Image: slide13_img11.png",{"_69":70,"_71":328,"_73":231,"_75":329},"5e0f339a-aac7-4992-ad26-7c55e6fd6cc8","photo: Image: slide13_img12.png",[],[],[333],{"_111":334,"_7":113,"_25":114,"_115":116,"_117":310,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":335},"V13_0_45802f2d",{"_135":116,"_117":310,"_118":119,"_120":121,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/13/master/c15ced2f-8abb-4453-bd99-5f6add51d257.webp",{"_21":339,"_27":340,"_29":341,"_31":342,"_33":313,"_35":157,"_36":179,"_38":343,"_42":344,"_44":345,"_46":346,"_48":347,"_50":351,"_52":352,"_54":353,"_55":157,"_56":57},{"_23":24,"_25":26},14,"b353d236-6bb0-495c-a1a5-593255393bf3","# Линейная регрессия\n### Преимущества\n\nМожет обучаться на малом наборе данных\nУстойчива к переобучению\nБыстрое обучение и применение\nМожет экстраполировать тренд\n### - факторы модели\n\\- параметры регрессии\nНедостатки\n\nНе может выявлять нелинейные взаимосвязи\nОбычно уступает по точности прогнозирования другим методам\n14",{"_16":315,"_40":316},[],[],[],[348],{"_111":349,"_7":113,"_25":114,"_115":116,"_117":340,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":350},"V14_0_45802f2d",{"_135":116,"_117":340,"_118":119,"_120":121,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/14/e59c2b6c-353d-43af-b681-6d74320d100d.webp","watermark_blur",{"_21":355,"_27":194,"_29":356,"_31":357,"_33":313,"_35":157,"_36":179,"_38":358,"_42":360,"_44":364,"_46":365,"_48":366,"_50":370,"_52":371,"_54":353,"_55":157,"_56":57},{"_23":24,"_25":26},"b60fab3c-0951-4222-8434-90dfb111f28f","# Логистическая регрессия\n### Это модель классификации, несмотря на название!\n### Преимущества\n\nМожет обучаться на малом наборе данных\nУстойчива к переобучению\nБыстрое обучение и применение\nМожет экстраполировать тренд\nНедостатки\n\nНе может выявлять нелинейные взаимосвязи\nОбычно уступает по точности прогнозирования другим методам\n15",{"_16":359,"_40":316},"Логистическая регрессия",[361],{"_69":70,"_71":362,"_73":231,"_75":363},"0962685a-203c-43aa-bee0-2c92a016d7d0","photo: Image: slide15_img13.png",[],[],[367],{"_111":368,"_7":113,"_25":114,"_115":116,"_117":194,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":369},"V15_0_45802f2d",{"_135":116,"_117":194,"_118":119,"_120":121,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/15/ffdd1814-2451-420a-bba3-2abce9fc8c4d.webp",{"_21":373,"_27":374,"_29":375,"_31":376,"_33":313,"_35":157,"_36":37,"_38":377,"_42":379,"_44":386,"_46":387,"_48":388,"_50":389,"_52":390,"_54":353,"_55":157,"_56":57},{"_23":24,"_25":26},16,"ea263e2c-aa5a-4c1a-a333-995ac8678374","# Дерево решений\n### Преимущества\n\nМожет обучаться на малом наборе данных\nМодель устойчива к переобучению\nХорошая интерпретируемость\nНедостатки\nНе может экстраполировать тренд\nУступает по точности прогнозирования другим методам\n16",{"_16":378,"_40":316},"Дерево решений",[380,383],{"_69":70,"_71":381,"_73":164,"_75":382},"d9dd33d8-19d2-4dfd-8d32-c941c5d10d25","diagram: A flowchart decision tree determining whether to grant a loan based on credit history and conditions.",{"_69":70,"_71":384,"_73":164,"_75":385},"46d65736-1be1-4644-8f3f-a497a20d6da9","diagram: A decision tree visualization from scikit-learn showing splits, entropy, samples, and values.",[],[],[],[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/16/ee45ab19-7615-4b5f-b074-c4188c1c4e49.webp",{"_21":392,"_27":393,"_29":394,"_31":395,"_33":313,"_35":157,"_36":37,"_38":396,"_42":399,"_44":403,"_46":404,"_48":405,"_50":409,"_52":410,"_54":353,"_55":157,"_56":57},{"_23":24,"_25":26},17,"1b2703ed-9847-42dc-a626-c9012be3853a","# Ансамблевые методы(Random forest)\n### Преимущества\n\nТочность выше чем у дерева решений\nОбучение и применение хорошо распараллеливается\nНедостатки\n\nНе может экстраполировать тренд\nМодель менее устойчива к переобучению, чем дерево решений\nСлабая интерпретируемость\n17",{"_16":397,"_40":398},"Ансамблевые методы(Random forest)","Random forest",[400],{"_69":70,"_71":401,"_73":164,"_75":402},"08144266-b0b0-4653-a999-d67368d36312","diagram: A workflow diagram representing an ensemble learning model combining multiple classifiers.",[],[],[406],{"_111":407,"_7":113,"_25":114,"_115":116,"_117":393,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":26,"_125":126,"_127":128,"_129":130,"_131":119,"_132":119,"_133":408},"V17_0_45802f2d",{"_135":116,"_117":393,"_118":119,"_120":121,"_129":130},[],"https://files.wonderslide.com/storage/be/53/2329/chats/5e17e82d-2a95-49c3-b876-ab2246e3163a/presentations/be5d9118-7649-434c-933c-1800cdfac725/slides/17/a96b0e55-860a-4fd6-9525-8f903c934dcd.webp","downloadUrl","actionData","errors"]