Level up your coding skills and quickly land a job. This is the best place to expand your knowledge and get prepared for your next interview.
LeetCode - The World's Leading Online Programming Learning Platform
Level up your coding skills and quickly land a job. This is the best place to expand your knowledge and get prepared for your next interview.
Home – Google Tech Dev Guide
Grow Your Technical Skills with Google Whether you're new to computer science or an experienced coder, there’s something for you here in Google’s Tech Dev Guide. We’ve carefully curated materials from various sources, including some made by Google, that you can use to grow your technical skills, supplement your coursework, and prepare for interviews. Interested in pursuing a career in business? Ch
TensorFlowã«ã‚ˆã‚‹ãƒ‡ã‚£ãƒ¼ãƒ—ラーニングã§ã€ã‚¢ã‚¤ãƒ‰ãƒ«ã®é¡”ã‚’è˜åˆ¥ã™ã‚‹ - ã™ãŽã‚ƒãƒ¼ã‚“メモ
以å‰ã¯ MNISTã®ä¾‹ã‚’使ã£ã¦ç”»åƒè˜åˆ¥ã‚’試ã—ã¦ã¿ãŸ ã‘ã©ã€æ¬¡ã¯ã‚«ãƒ©ãƒ¼ç”»åƒã«ã¤ã„ã¦ã®è˜åˆ¥ã‚’試ã—ã¦ã¿ã‚‹ã€‚ 「アイドルãªã‚“ã¦ã¿ã‚“ãªåŒã˜é¡”ã«è¦‹ãˆã‚‹ã€ã£ã¦ 最近も言ã‚ã‚Œã¦ã‚‹ã®ã‹ã©ã†ã‹çŸ¥ã‚‰ãªã„ã‘ã©ã€è‡ªåˆ†ã‚‚ã¤ã„5å¹´å‰ãらã„ã¾ã§ã¯ãã†æ€ã£ã¦ã„ãŸã‚ã‘ã§ã€‚ãã®è˜åˆ¥ã‚’機械å¦ç¿’ã§ã‚„ã£ã¦ã¿ã‚ˆã† ã¨ã„ã†è©¦ã¿ã€‚ 最近ã¯ã»ã¨ã‚“ã©ãƒ©ã‚¤ãƒ–ã«è¡Œã‹ãªããªã£ã¦ã—ã¾ã£ãŸã‘ã©å¤§å¥½ããªã‚‚ã‚‚ã„ã‚クãƒãƒ¼ãƒãƒ¼Zã¡ã‚ƒã‚“ã‚’é¡Œæã«ã—ã¦ã¿ã‚‹ã“ã¨ã«ã€‚ 5人ã®ãƒ¡ãƒ³ãƒãƒ¼ã®é¡”ã¯æ©Ÿæ¢°å¦ç¿’ã«ã‚ˆã£ã¦ã©ã‚Œãらã„分類ã§ãるよã†ã«ãªã‚‹ã®ã‹ï¼Ÿï¼Ÿ CIFAR-10 CIFAR-10 ã¨ã„ã†ã€32×32サイズã®ã‚«ãƒ©ãƒ¼ç”»åƒã‚’10種類ã®ã‚¯ãƒ©ã‚¹ã«åˆ†é¡žã™ã‚‹è˜åˆ¥èª²é¡ŒãŒã‚ã‚Šã€ãã®ãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆãŒå…¬é–‹ã•ã‚Œã¦ã„る。ã“れを実際ã«TensorFlowã§å¦ç¿’ã™ã‚‹ãŸã‚ã®ç•³ã¿è¾¼ã¿ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®ãƒ¢ãƒ‡ãƒ«ã‚„関数ãªã©ãŒtensorflow.models.image.cifar10パッケージã«åŒæ¢±ã•ã‚Œã¦
ディープラーニングã®æœ‰åライブラリ5種を最çŸè·é›¢ã§è©¦ã™åŠæ—¥ã‚³ãƒ¼ã‚¹ï¼ˆTensorFlow, Chainer, Caffe, DeepDream, 画風変æ›ï¼‰ - ãã®å¾Œã®ãã®å¾Œ
「ã„ã¤ã‹å‹‰å¼·ã—よã†ã€ã¨äººå·¥çŸ¥èƒ½ï¼æ©Ÿæ¢°å¦ç¿’ï¼ãƒ‡ã‚£ãƒ¼ãƒ—ラーニング（Deep Learning）ã¨ã„ã£ãŸãƒˆãƒ”ックã®è¨˜äº‹ã®è¦‹ã¤ã‘ã¦ã¯ã‚¢ãƒ¼ã‚«ã‚¤ãƒ–ã—ã¦ããŸã‚‚ã®ã®ã€çµå±€2015å¹´ã¯ä½•ä¸€ã¤ã‚„らãšã«çµ‚ã‚ã£ã¦ã—ã¾ã£ãŸã®ã§ã€ã¨ã«ã‹ã一æ©ã§ã‚‚足をè¸ã¿å‡ºã™ã¹ãã€æœ¬è³ªçš„ãªç†è§£ç‰ã¯ã•ã¦ãŠãã€ã¨ã«ã‹ã試ã—ã¦ã¿ã‚‹ã¨ã„ã†ã“ã¨ã‚’ã‚„ã£ã¦ã¿ã¾ã—ãŸã€‚ 試ã—ãŸã®ã¯ã€TensorFlowã€Chainerã€Caffe ã¨ã„ã£ãŸæ©Ÿæ¢°å¦ç¿’ãŠã‚ˆã³ãƒ‡ã‚£ãƒ¼ãƒ—ラーニングã®ä»£è¡¨çš„ãªãƒ©ã‚¤ãƒ–ラリï¼ãƒ•ãƒ¬ãƒ¼ãƒ ワーク3種ã¨ã€2015å¹´ã«è©±é¡Œã«ãªã£ãŸãƒ‡ã‚£ãƒ¼ãƒ—ラーニングを利用ã—ãŸã‚¢ãƒ—リケーション2種（DeepDreamã€chainer-gogh）。 （DeepDreamã§è©¦ã—ãŸçµæžœç”»åƒï¼‰ タイトルã«åŠæ—¥ã¨æ›¸ãã¾ã—ãŸãŒã€ãŸã¨ãˆã°TensorFlowã¯ç’°å¢ƒæ§‹ç¯‰ã ã‘ãªã‚‰10分もã‚ã‚Œã°çµ‚ã‚ã‚‹ã§ã—ょã†ã—ã€Chainerãªã‚“ã¦ã‚³ãƒžãƒ³ãƒ‰ä¸€ç™ºãªã®ã§5秒ãらã„ã§ã™ã€‚Caffeã¯åƒ•ã¯ãƒ
æ„訳：2009å¹´ã‹ã‚‰2014å¹´ã®æ©Ÿæ¢°å¦ç¿’ã®ç†è«–的進æ©ã«ã¤ã„ã¦æ•™ãˆã¦ãã ã•ã„ - Dust Theory
Quoraã«æŠ•ç¨¿ã•ã‚ŒãŸè³ªå•Â "What has happened in theoretical machine learning in the last 5 years (2009-2014)?" 「2009å¹´ã‹ã‚‰2014å¹´ã®æ©Ÿæ¢°å¦ç¿’ã®ç†è«–的進æ©ã«ã¤ã„ã¦æ•™ãˆã¦ãã ã•ã„〠ã¸ã®ã€æ©Ÿæ¢°å¦ç¿’ç ”ç©¶è€…Yisong Yueã®å›žç”を翻訳ã—ã¾ã—ãŸã€‚ãã®ã¾ã¾ã ã¨é€šã˜ãªã„部分ã¯æ„訳ã—ã¦ã„ã‚‹ã®ã§ã€åŽŸæ–‡ã«å¿ 実ã§ã¯ã‚ã‚Šã¾ã›ã‚“。 「2009å¹´ã‹ã‚‰2014å¹´ã®æ©Ÿæ¢°å¦ç¿’ã®ç†è«–的進æ©ã«ã¤ã„ã¦æ•™ãˆã¦ãã ã•ã„ã€ã¸ã®å›žç” ç§ã®å€‹äººçš„ãªè¦–点ã§è©±ã—ã¾ã™ã€‚ 潜在変数モデルã«å¯¾ã™ã‚‹æœ€é©æŽ¨å®šã«ã¤ã„ã¦: ç†è«–çš„ãªé€²æ©ã¨ã—ã¦ç§ãŒæœ€åˆã«æ€ã„ã¤ãã®ã¯ã€æ½œåœ¨å¤‰æ•°ãƒ¢ãƒ‡ãƒ«ã«å¯¾ã™ã‚‹ï¼ˆã»ã¼ï¼‰æœ€é©ãªæŽ¨å®šã§ã™ã€‚ã“ã®æ½œåœ¨å¤‰æ•°ãƒ¢ãƒ‡ãƒ«ã®æœ€é©æŽ¨å®šã¯ä¸€èˆ¬çš„ã«éžå‡¸ãªå•é¡Œã§ã‚ã‚Šã€ã¤ã¾ã‚Šï¼ˆã‚ˆã知られã¦ã„る）凸最é©åŒ–ã®æ‰‹æ³•ã®é€šç”¨ã—ãªã„難題ã§ã™ã€‚ 最もè¼ã‹ã—ã„アプãƒãƒ¼ãƒã¯ã‚¹ãƒšã‚¯ãƒˆ
1
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
{{#tags}}- {{label}}
{{/tags}}