投稿時間:2022-12-27 14:25:56 RSSフィード2022-12-27 14:00 分まとめ(30件)

カテゴリー等	サイト名等	記事タイトル・トレンドワード等	リンクURL	頻出ワード・要約等/検索ボリューム	登録日
ROBOT	ロボスタ	定時になると「あるくメカトロウィーゴ」が一斉に踊り出す　あぶくま信用金庫が合計10台のウィーゴを設置	https://robotstart.info/2022/12/27/mechatrowego-dance-service.html	定時になると「あるくメカトロウィーゴ」が一斉に踊り出すあぶくま信用金庫が合計台のウィーゴを設置シェアツイートはてブ株式会社リビングロボットは同社が開発した「あるくメカトロウィーゴ」をあぶくま信用金庫の営業店窓口に設置したことを発表した。	2022-12-27 04:13:54
IT	ITmedia 総合記事一覧	[ITmedia ビジネスオンライン] EVを買う際、希望するメーカーランキング　3位「ホンダ」、2位「日産」　納得の1位は？	https://www.itmedia.co.jp/business/articles/2201/02/news029.html	itmedia	2022-12-27 13:45:00
IT	ITmedia 総合記事一覧	[ITmedia News] “1人1台”で放置される学校の「コンピュータ教室」　それでも文科省が残したい理由	https://www.itmedia.co.jp/news/articles/2212/27/news103.html	itmedia	2022-12-27 13:30:00
IT	ITmedia 総合記事一覧	[ITmedia ビジネスオンライン] あらゆるジェンダーの新成人を祝う「SEIJIN-SHIKI」開催　新規顧客へ訴求	https://www.itmedia.co.jp/business/articles/2212/27/news108.html	itmedia	2022-12-27 13:03:00
python	Pythonタグが付けられた新着投稿 - Qiita	PythonでDeePL APIの英訳結果を、文字列（str）型として受け取る方法	https://qiita.com/komeda_coffee_24_sotu/items/a10e4ee34087a5104d2a	deepl	2022-12-27 13:45:52
python	Pythonタグが付けられた新着投稿 - Qiita	【Python】特定の宛先からメールが来たら通知に出すようにしてみた	https://qiita.com/tamago_pinoko/items/a1f0897fb8eef1336c91	googlecloudplatf	2022-12-27 13:05:26
技術ブログ	Developers.IO	AWS App Runner を動かしてみたくてワークショップをやってみた	https://dev.classmethod.jp/articles/try-aws-app-runner-workshop/	apprunner	2022-12-27 04:47:38
技術ブログ	Developers.IO	【レポート】A dynamic opportunity: Understand AWS China and grow your business #ENT230 #reinvent	https://dev.classmethod.jp/articles/reinvent2022-ent230-aws-china/	awschina	2022-12-27 04:43:36
技術ブログ	Developers.IO	1年使った検証用個人AWSアカウント内のリソースを大掃除してみる	https://dev.classmethod.jp/articles/1year-test-aws-account-clean-up/	都度	2022-12-27 04:41:59
海外TECH	DEV Community	Gender Determination with Morphometry of Eyes using deep learning model.	https://dev.to/anurag629/gender-determination-with-morphometry-of-eyes-using-deep-learning-model-5f83	Gender Determination with Morphometry of Eyes using deep learning model In this tutorial we will be using deep learning techniques to determine the gender of a person based on the morphometry of their eyes We will start by importing the necessary modules including TensorFlow Keras and Matplotlib Next we will define some variables such as image size batch size and the number of channels We will then use the image dataset from directory function to load our dataset This function allows us to easily load images from a directory and organize them into a dataset that we can use for training and testing our model We will also visualize some of the images in the dataset to get a better understanding of what we are working with After loading the dataset we will split it into train validation and test sets using a custom function that we define This is a crucial step as we will use the train set for training our model the validation set for evaluating the model during training and the test set for evaluating the model s performance on unseen data Next we will prepare our dataset for training by applying some transformations such as shuffling caching and prefetching This helps to improve the efficiency and speed of our training process We will then build our model using the Keras Sequential API This allows us to easily stack layers of different types to create our model In this case we will use a combination of convolutional and dense layers After building the model we will compile it by specifying the loss function and optimization algorithm that we want to use We will also specify some metrics that we want to track during training With our model compiled we are now ready to start training We will do this by calling the fit function and passing it to our training and validation datasets as well as the number of epochs that we want to train for Finally we will evaluate the model s performance on the test set and visualize the results using Matplotlib By following these steps we should be able to create a model that can accurately determine the gender of a person based on the morphometry of their eyes Here is an example of how you can classify Gender with Morphometry of Eyes using a deep learning model in Python using the TensorFlow library Importing modulesimport tensorflow as tffrom tensorflow keras import models layersimport matplotlib pyplot as pltfrom tensorflow import kerasprint Done Defining variablesimage size batch size channels epoches print Done loading dataset using image datset from directoryimage dataset from directory is a function in TensorFlow that can be used to create a tf data Dataset object for images stored in a directory structure It can be used to load and preprocess the images for training evaluation or prediction dataset tf keras preprocessing image dataset from directory kaggle input gender determination with morphometry of eyes train shuffle True image size image size image size batch size batch size print Done Printing class namesclass names dataset class namesclass namesprint Done visualizing imagesplt figure figsize for image batch label batch in dataset take for i in range ax plt subplot i plt imshow image batch i numpy astype uint plt title class names label batch i print Done splitting our dataset into train val and testdef get dataset partitions tf ds train split val split test split shuffle True shuffle size ds size len ds if shuffle ds ds shuffle shuffle size seed train size int train split ds size val size int val split ds size train ds ds take train size val ds ds skip train size take val size test ds ds skip train size skip val size return train ds val ds test dsprint Done train ds val ds test ds get dataset partitions tf dataset print Done print Len train set len train ds print Len val set len val ds print Len test set len test ds print Done train ds train ds cache shuffle prefetch buffer size tf data AUTOTUNE val ds val ds cache shuffle prefetch buffer size tf data AUTOTUNE test ds test ds cache shuffle prefetch buffer size tf data AUTOTUNE print Done resizing and rescaling the imageresize and rescale tf keras Sequential layers experimental preprocessing Resizing image size image size layers experimental preprocessing Rescaling print Done applying data augmentationdata augumentation tf keras Sequential layers experimental preprocessing RandomFlip horizontal and vertical layers experimental preprocessing RandomRotation print Done building our model Load a pre trained modelbase model keras applications ResNet weights imagenet input shape Remove the final layers of the modelbase model layers pop base model layers pop Add your own layersmodel keras Sequential resize and rescale data augumentation base model keras layers Flatten keras layers Dense activation relu keras layers Dense activation sigmoid Freeze the weights of the pre trained modelfor layer in base model layers layer trainable False Compile the modelmodel compile optimizer adam loss binary crossentropy metrics accuracy print First fit started Train the model using the pre trained model as a starting pointmodel fit train ds batch size epochs validation data val ds print First fit done Fine tune the model by unfreezing some layersfor layer in base model layers layer trainable True Recompile the modelmodel compile optimizer adam loss binary crossentropy metrics accuracy print First fit started Continue training the modelhistory model fit train ds batch size epochs validation data val ds verbose print First fit done model summary print Done Plotting training and validation accuracy and loss Plot the training and validation accuracyplt plot history history accuracy plt plot history history val accuracy plt title Model accuracy plt ylabel Accuracy plt xlabel Epoch plt legend Train Test loc upper left plt show Plot the training and validation lossplt plot history history loss plt plot history history val loss plt title Model loss plt ylabel Loss plt xlabel Epoch plt legend Train Test loc upper left plt show Now we will test the model for the given test data set Set the image size and batch sizeIMAGE SIZE BATCH SIZE Create an ImageDataGenerator object for the imagesimage generator ImageDataGenerator rescale Load the images from the folderdata generator image generator flow from directory kaggle input gender determination with morphometry of eyes test target size IMAGE SIZE batch size BATCH SIZE class mode None Do not generate labels shuffle False Do not shuffle the images Predict the class of each imagepredictions model predict data generator steps len data generator Get the class labelsclass labels data generator class indicesclass labels v k for k v in class labels items Invert the dictionary Print the predictionsfor i prediction in enumerate predictions print f Image i class labels np argmax prediction Thanks for liking it	2022-12-27 04:23:57
金融	ニッセイ基礎研究所	人口構成からみる社会の変化(2)－性・年齢階層別人口の前後５年間の変化－	https://www.nli-research.co.jp/topics_detail1/id=73379?site=nli	ー都道府県別の差異年から年の変化全体での増加率が高い高齢層について都道府県別にみると、歳以上の増加率は埼玉県、千葉県、神奈川県、愛知県、大阪府で割を超えて高く、特に埼玉県の男性ではと増加率の高さが目立っている。	2022-12-27 13:57:59
金融	ニッセイ基礎研究所	コロナ禍を受けた改正感染症法はどこまで機能するか－医療機関と都道府県による事前協定制度などの行方を考える	https://www.nli-research.co.jp/topics_detail1/id=73377?site=nli	例えば、先に触れた地域医療構想について見ると、都道府県を中心に医療機関の経営者や市町村の関係者、介護事業所の経営者、住民などで構成する「地域医療構想調整会議」での合意形成と自主的な対応が想定されており、国や都道府県は民間医療機関に対して病床削減などを命令できない。	2022-12-27 13:30:27
金融	日本銀行：RSS	基調的なインフレ率を捕捉するための指標	http://www.boj.or.jp/research/research_data/cpi/index.htm	捕捉	2022-12-27 14:00:00
海外ニュース	Japan Times latest articles	Japan’s reconstruction minister resigns following funding scandal	https://www.japantimes.co.jp/news/2022/12/27/national/politics-diplomacy/kenya-akiba-resignation/	Japan s reconstruction minister resigns following funding scandalKenya Akiba has admitted to having two political groups affiliated with him pay million to his mother and his wife as “rent between	2022-12-27 13:33:41
海外ニュース	Japan Times latest articles	Battling the odds to rise to the top: One woman’s career in the IT sector	https://www.japantimes.co.jp/news/2022/12/27/business/women-work-shingu/	Battling the odds to rise to the top One woman s career in the IT sectorYuki Shingu found taking a career break to help nurse her ailing father gave her a broader perspective on her rise through company ranks	2022-12-27 13:13:14
ニュース	BBC News - Home	Wallasey pub shooting: Two arrested after Elle Edwards killed on Christmas Eve	https://www.bbc.co.uk/news/uk-england-merseyside-64099145?at_medium=RSS&at_campaign=KARANGA	edwards	2022-12-27 04:31:41
ニュース	BBC News - Home	US winter storm traps New York State residents in cars	https://www.bbc.co.uk/news/world-us-canada-64099642?at_medium=RSS&at_campaign=KARANGA	canada	2022-12-27 04:55:16
ニュース	BBC News - Home	China ends quarantine to begin reopening in January	https://www.bbc.co.uk/news/world-asia-china-64097497?at_medium=RSS&at_campaign=KARANGA	covid	2022-12-27 04:13:42
ニュース	BBC News - Home	South Korea lifts ban on import of adult sex dolls	https://www.bbc.co.uk/news/business-64099652?at_medium=RSS&at_campaign=KARANGA	dolls	2022-12-27 04:55:18
ニュース	BBC News - Home	The Papers: 'Light of our life' and '£40,000 hit to incomes'	https://www.bbc.co.uk/news/blogs-the-papers-64099075?at_medium=RSS&at_campaign=KARANGA	front	2022-12-27 04:11:24
北海道	北海道新聞	秋葉復興相を更迭、閣僚４人目　後任に渡辺氏再登板、政権打撃	https://www.hokkaido-np.co.jp/article/781229/	岸田文雄	2022-12-27 13:32:32
北海道	北海道新聞	杉田水脈政務官が辞表提出　性的少数者巡り差別的言動	https://www.hokkaido-np.co.jp/article/781253/	性的少数者	2022-12-27 13:32:00
北海道	北海道新聞	パナ、炊飯器国内生産終了へ　２３年６月、中国に移管	https://www.hokkaido-np.co.jp/article/781224/	移管	2022-12-27 13:16:28
北海道	北海道新聞	外国人差別で人権救済申し立て　生活保護巡る安城市職員の発言	https://www.hokkaido-np.co.jp/article/781252/	人権救済申し立て	2022-12-27 13:31:00
北海道	北海道新聞	オマーンとＬＮＧ契約へ　年２００万トン、長期調達に合意	https://www.hokkaido-np.co.jp/article/781250/	三井物産	2022-12-27 13:26:00
北海道	北海道新聞	李明博元大統領の恩赦決定、韓国　残り１５年の刑期免除	https://www.hokkaido-np.co.jp/article/781242/	元大統領	2022-12-27 13:12:06
ニュース	Newsweek	ロヒンギャ難民船がインドネシアに相次ぎ漂着　ミャンマー脱出後各地で苦難続く	https://www.newsweekjapan.jp/stories/world/2022/12/post-100463.php	しかしマレーシア政府はコロナ感染防止やロヒンギャ難民の中にARSAなどの武装組織メンバーが混入している可能性があることなどから多くの難民船を食料や飲料水を与えたうえで国際海域に追い返していた。	2022-12-27 13:40:39
IT	週刊アスキー	パン好きのパン好きによるパン好きのための祭典！　「パンのフェス2023春 in 横浜赤レンガ」2023年3月3日～ 5日開催	https://weekly.ascii.jp/elem/000/004/118/4118977/	横浜赤レンガ倉庫	2022-12-27 13:50:00
IT	週刊アスキー	『ロードス島戦記（コンシューマー版）』が「プロジェクトEGG」で本日12月27日から配信開始！	https://weekly.ascii.jp/elem/000/004/118/4118988/	配信サービス	2022-12-27 13:15:00
IT	週刊アスキー	キャンペーンキャラクターは髙橋ひかるさんとウルトラマン　「はたちの献血」キャンペーン2023年1月1日～2月28日開催	https://weekly.ascii.jp/elem/000/004/118/4118967/	日本赤十字	2022-12-27 13:10:00