{"id":1862,"date":"2021-11-29T17:06:48","date_gmt":"2021-11-29T17:06:48","guid":{"rendered":"https:\/\/eic.cefet-rj.br\/~ebezerra\/?page_id=1862"},"modified":"2022-03-17T17:22:44","modified_gmt":"2022-03-17T17:22:44","slug":"dl_2021","status":"publish","type":"page","link":"https:\/\/eic.cefet-rj.br\/~ebezerra\/dl_2021\/","title":{"rendered":"GCC1917 &#8211; T\u00f3picos Especiais em Programa\u00e7\u00e3o I (2021.2)"},"content":{"rendered":"<p style=\"text-align: center;\">\n<\/p><h3>Locais e hor\u00e1rios das aulas<\/h3>\n<ul>\n<li>5as-feiras, das 14:00h \u00e0s 17:00h, por videoconfer\u00eancia (plataforma MS Teams)<\/li>\n<\/ul>\n<hr>\n<h3>Objetivos<\/h3>\n<p>Apresentar uma introdu\u00e7\u00e3o \u00e0 t\u00e9cnicas para treinamento de redes neurais profundas.<\/p>\n<hr>\n<h3>Ementa<\/h3>\n<p>Introdu\u00e7\u00e3o ao numpy e ao PyTorch; Conceito de tensor; Regress\u00e3o linear, Regress\u00e3o log\u00edstica; Fun\u00e7\u00f5es de ativa\u00e7\u00e3o e camadas ocultas; Uso de GPUs; Redes neurais convolucionais; Redes neurais recorrentes; Autoencoders; Graph Neural Nets; Transformers.<\/p>\n<p>Veja tamb\u00e9m o <a href=\"https:\/\/eic.cefet-rj.br\/portal\/wp-content\/uploads\/GCC1917.pdf\">plano de ensino<\/a> da disciplina.<\/p>\n<hr>\n<h3><span style=\"color: inherit; font-size: 1.56em;\">Livros-textos<\/span><\/h3>\n\n\n<figure class=\"wp-block-gallery columns-2 is-cropped wp-block-gallery-1 is-layout-flex wp-block-gallery-is-layout-flex\"><ul class=\"blocks-gallery-grid\"><li class=\"blocks-gallery-item\"><figure><a href=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-with-pytorch-cover-2.jpg\"><img loading=\"lazy\" decoding=\"async\" width=\"283\" height=\"354\" src=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-with-pytorch-cover-2.jpg\" alt=\"\" data-id=\"1877\" data-full-url=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-with-pytorch-cover-2.jpg\" data-link=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/dl_2021\/dl-with-pytorch-cover-2\/\" class=\"wp-image-1877\" srcset=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-with-pytorch-cover-2.jpg 283w, https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-with-pytorch-cover-2-240x300.jpg 240w\" sizes=\"auto, (max-width: 283px) 100vw, 283px\" \/><\/a><\/figure><\/li><li class=\"blocks-gallery-item\"><figure><a href=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-book-cover-1.jpg\"><img loading=\"lazy\" decoding=\"async\" width=\"263\" height=\"349\" src=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-book-cover-1.jpg\" alt=\"\" data-id=\"1878\" data-full-url=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-book-cover-1.jpg\" data-link=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/dl_2021\/dl-book-cover-1\/\" class=\"wp-image-1878\" srcset=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-book-cover-1.jpg 263w, https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2021\/11\/dl-book-cover-1-226x300.jpg 226w\" sizes=\"auto, (max-width: 263px) 100vw, 263px\" \/><\/a><\/figure><\/li><\/ul><\/figure>\n\n\n<hr \/>\n<h3>Aulas<\/h3>\n<p>Veja tamb\u00e9m o <a href=\"https:\/\/www.dropbox.com\/s\/qftweptnx2agp8o\/GCC1917-PlanoCurso-2021.2.xlsx?dl=0\" target=\"_blank\" rel=\"noopener\">plano de curso<\/a> da disciplina e o <a href=\"http:\/\/www.cefet-rj.br\/attachments\/article\/199\/MAR%20-%20CAL%20GRAD%20MAR%202021%20-%20V3%20(out%202021).pdf\">calend\u00e1rio acad\u00eamico das gradua\u00e7\u00f5es<\/a> do CEFET\/RJ.<\/p>\n<table style=\"border: 1px solid #cccccc;\" width=\"100%\">\n<tbody>\n<tr>\n<td style=\"border: 1px solid #cccccc;\"><b>Aula<\/b><\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\"><b>Data<\/b><\/td>\n<td style=\"border: 1px solid #cccccc;\"><b>T\u00f3pico<\/b><\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">1<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">25\/nov<\/td>\n<td style=\"border: 1px solid #cccccc;\">Log\u00edstica do curso, plano de aulas (<a href=\"https:\/\/www.dropbox.com\/s\/8ckghmz7i43eepe\/DL00%20%28Apresenta%C3%A7%C3%A3o%20do%20curso%29.pptx?dl=0\" target=\"_blank\" rel=\"noopener\">slides<\/a>)<br \/>Redes Neurais Profundas: O que s\u00e3o? Como vivem? De que se alimentam?\u00a0(<a href=\"https:\/\/youtu.be\/etebnvblYLY\">video<\/a>,\u00a0<a href=\"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-content\/uploads\/2018\/02\/2020-UFF-IC-RedesNeuraisProfundas.pdf\">slides<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">2<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">02\/dez<\/td>\n<td style=\"border: 1px solid #cccccc;\">numpy, PyTorch, regress\u00e3o linear (<a href=\"https:\/\/colab.research.google.com\/drive\/1v8XMeD93jZhGvkEo1OzMErMTirevxd3I?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">3<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">09\/dez<\/td>\n<td style=\"border: 1px solid #cccccc;\">Classes Dataset e DataLoader (<a href=\"https:\/\/colab.research.google.com\/drive\/1MztEmdzyVhjwBaeFxKjlYIkaY1wXoaSe?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">4<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">16\/dez<\/td>\n<td style=\"border: 1px solid #cccccc;\">Regress\u00e3o log\u00edstica, softmax, entropia cruzada (<a href=\"https:\/\/colab.research.google.com\/drive\/1MztEmdzyVhjwBaeFxKjlYIkaY1wXoaSe?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">5<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">23\/dez<\/td>\n<td style=\"border: 1px solid #cccccc;\">Fun\u00e7\u00f5es de ativa\u00e7\u00e3o, camadas ocultas (<a href=\"https:\/\/drive.google.com\/file\/d\/1OPs5D3nggPbRjoV7s2Ycp3I6oDvWA_hM\/view?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">6<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">30\/dez<\/td>\n<td style=\"border: 1px solid #cccccc;\">Uso de GPUs (<a href=\"https:\/\/drive.google.com\/file\/d\/1OPs5D3nggPbRjoV7s2Ycp3I6oDvWA_hM\/view?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">7<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">13\/jan<\/td>\n<td style=\"border: 1px solid #cccccc;\">Redes neurais convolucionais (<a href=\"https:\/\/colab.research.google.com\/drive\/1qbZ-XbhdjoBuJfcTSsnUNxDOWKuB7-a2?usp=sharing\">notebook<\/a>, <a href=\"https:\/\/www.dropbox.com\/s\/c4i8qsgz75k3id5\/convnets-1a-parte.pptx?dl=0\">slides<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">8<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">27\/jan<\/td>\n<td style=\"border: 1px solid #cccccc;\">Previs\u00e3o de s\u00e9ries temporais com Conv1d (<a href=\"https:\/\/colab.research.google.com\/drive\/1RYPQIQ8v5ue5_T7-1fwvXOfWu46FqBzy?usp=sharing\">notebook<\/a>, <a href=\"https:\/\/www.dropbox.com\/s\/osgq5f3v3zyyp9q\/convnets-2a-parte.pptx?dl=0\">slides<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">9<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">03\/fev<\/td>\n<td style=\"border: 1px solid #cccccc;\">Aspectos pr\u00e1ticos do treinamento &#8211; 1a parte (<a href=\"https:\/\/www.dropbox.com\/s\/vtlpkqddcx1n148\/Treinamento-aspectos-praticos-parte1.pptx?dl=0\">slides<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">10<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">10\/fev<\/td>\n<td style=\"border: 1px solid #cccccc;\">Redes neurais recorrentes &#8211; 1a parte (<a href=\"https:\/\/www.dropbox.com\/s\/u8wuwzi4vzq0qis\/RedesNeuraisRecorrentes.pptx?dl=0\">slides<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">11<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">17\/fev<\/td>\n<td style=\"border: 1px solid #cccccc;\">Regulariza\u00e7\u00e3o: early stopping e dropout (<a href=\"https:\/\/www.dropbox.com\/s\/fcn9aihqlgql2qe\/Early-stoppin-n-dropout.pptx?dl=0\">slides<\/a>, <a href=\"https:\/\/colab.research.google.com\/drive\/1NrZ3gPIe04YYvsV2ooQsjfe35vL8QdTr?usp=sharing\">notebook<\/a>)<br \/>Aplica\u00e7\u00f5es (<a href=\"https:\/\/www.dropbox.com\/s\/88zrr99634imibo\/Apps-n-final-remarks.pptx?dl=0\">slides<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">12<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">24\/fev<\/td>\n<td style=\"border: 1px solid #cccccc;\">Augusto Fonseca: Redes neurais para grafos (<a href=\"https:\/\/www.dropbox.com\/s\/v8t7t5z0q3wp5qs\/gnn.pdf?dl=0\">slides<\/a>, <a href=\"https:\/\/colab.research.google.com\/drive\/1JYhXYgYWDkh3ZhvBMjnrE-sliWTNxxub?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">13<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">03\/mar<\/td>\n<td style=\"border: 1px solid #cccccc;\">Redes neurais recorrentes &#8211; 2a parte (<a href=\"https:\/\/colab.research.google.com\/drive\/1UBvVNV1yYtOU0lPkJq0r4VtSqBD4keMi?usp=sharing\">notebook<\/a>); <br \/>Autoencoders (<a href=\"https:\/\/www.dropbox.com\/s\/j2ssiw460eiodyy\/RedesAutocodificadoras.pptx?dl=0\">slides<\/a>, <a href=\"https:\/\/colab.research.google.com\/drive\/1ejtDiouPkjKceIi9zvxTob6UJ9KjzVqI?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">14<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">10\/mar<\/td>\n<td style=\"border: 1px solid #cccccc;\">GANs (<a href=\"https:\/\/www.dropbox.com\/s\/fz1jzonxx33zrwy\/GANs.pptx?dl=0\">slides<\/a>, <a href=\"https:\/\/colab.research.google.com\/drive\/1OaOkXw1x6IHQdXP1E2RYFrXfXwjgAeJX?usp=sharing\">notebook<\/a>); Aspectos pr\u00e1ticos do treinamento &#8211; 2a parte (<a href=\"https:\/\/colab.research.google.com\/drive\/1S9lcx-dX3ckfENedvbDq7l1nMBx5N6Xk?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">15<\/td>\n<td style=\"border: 1px solid #cccccc; text-align: center;\">17\/mar<\/td>\n<td style=\"border: 1px solid #cccccc;\">NLP &amp; Transformers (<a href=\"https:\/\/www.dropbox.com\/s\/ecf74f5a4z6qfba\/NLP-n-Transformers.pptx?dl=0\">slides<\/a>, <a href=\"https:\/\/colab.research.google.com\/drive\/1QyAkexwJfEeCAa30AK2brzdLioe4hhVG?usp=sharing\">notebook<\/a>)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<hr size=\"4\" width=\"100%\" \/>\n<p><!--\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<h3>Resumos<\/h3>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nOs resumos de palestras e artigos devem ter no m\u00e1ximo uma p\u00e1gina e devem identificar as principais ideias apresentadas em cada palestra. O objetivo dos resumos \u00e9 ajudar voc\u00ea a refletir sobre as ideias transmitidas pelo autor da palestra.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<h5>Resumo 1 - <span style=\"color: #ff0000; font-size: xx-small;\">Entrega a definir<\/span><\/h5>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAssistir ao TED Talk <a href=\"https:\/\/www.ted.com\/talks\/nick_bostrom_what_happens_when_our_computers_get_smarter_than_we_are\">What happens when our computers get smarter than we are?<\/a> de NICK BOSTROM.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<h5>Resumo 2 - <span style=\"color: #ff0000; font-size: xx-small;\">Entrega a definir<\/span><\/h5>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAssistir \u00e0 palestra <a href=\"https:\/\/www.youtube.com\/watch?v=B8J4uefCQMc\">The Master Algorithm<\/a> de Pedro Domingos. Os slides dessa palestra: <a href=\"&quot;https:\/\/www.slideshare.net\/SessionsEvents\/pedro-domingos-professor-university-of-washington-at-mlconf-atl-91815&lt;\/a\">.<\/a>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<h5>Resumo 3 - <span style=\"color: #ff0000; font-size: xx-small;\">Entrega a definir<\/span><\/h5>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAssistir ao TED Talk <a href=\"https:\/\/www.ted.com\/talks\/fei_fei_li_how_we_re_teaching_computers_to_understand_pictures\">How we're teaching computers to understand pictures<\/a> da Fei-Fei Li.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<h5>Resumo 4 - <span style=\"color: #ff0000; font-size: xx-small;\">Entrega a definir<\/span><\/h5>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nAssistir ao TED Talk <a href=\"https:\/\/www.ted.com\/talks\/sebastian_thrun_google_s_driverless_car\">Google's Driverless Car<\/a> de Sebastian Thrun.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<hr size=\"4\" width=\"100%\" \/>\n\n--><\/p>\n<p>\u00a0<\/p>\n<h3>Trabalhos pr\u00e1ticos<\/h3>\n<ul>\n<li><strong><a href=\"https:\/\/www.dropbox.com\/s\/7cbxz89dc14s9ph\/DL_T1.pdf?dl=0\">T1<\/a> <\/strong>&#8211; <span style=\"color: #ff0000;\">Entrega: 15\/fev<\/span><\/li>\n<li><strong><a href=\"https:\/\/www.dropbox.com\/s\/7fzcj1dbc0oblxr\/DL_T2.pdf?dl=0\">T2<\/a> &#8211;<\/strong>\u00a0<span style=\"color: #ff0000;\">Entrega: 21\/mar<\/span><\/li>\n<\/ul>\n<p>Para implementar os trabalhos, voc\u00ea ir\u00e1 precisar dominar os fundamentos da linguagem Python. Para isso, recomendo dois livros, ambos gratuitamente dispon\u00edveis:<\/p>\n<ol>\n<li><a href=\"https:\/\/automatetheboringstuff.com\">Automate the Boring Stuff with Python<\/a><\/li>\n<li><a href=\"https:\/\/drive.google.com\/file\/d\/10KIvvyvY4J17PhmzoWFR0qijzd4deuFl\/view\">Introdu\u00e7\u00e3o \u00e0 Programa\u00e7\u00e3o com Python<\/a><\/li>\n<\/ol>","protected":false},"excerpt":{"rendered":"<p>Locais e hor\u00e1rios das aulas 5as-feiras, das 14:00h \u00e0s 17:00h, por videoconfer\u00eancia (plataforma MS Teams) Objetivos Apresentar uma introdu\u00e7\u00e3o \u00e0 t\u00e9cnicas para treinamento de redes neurais profundas. Ementa Introdu\u00e7\u00e3o ao numpy e ao PyTorch; Conceito de tensor; Regress\u00e3o linear, Regress\u00e3o log\u00edstica; Fun\u00e7\u00f5es de ativa\u00e7\u00e3o e camadas ocultas; Uso de GPUs; Redes neurais convolucionais; Redes neurais [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-1862","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/pages\/1862","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/comments?post=1862"}],"version-history":[{"count":45,"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/pages\/1862\/revisions"}],"predecessor-version":[{"id":2045,"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/pages\/1862\/revisions\/2045"}],"wp:attachment":[{"href":"https:\/\/eic.cefet-rj.br\/~ebezerra\/wp-json\/wp\/v2\/media?parent=1862"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}