python |
Pythonタグが付けられた新着投稿 - Qiita |
seleniumを利用してGoogle認証を通す方法 |
https://qiita.com/std-flower/items/ec59a46b7ad0e5d225ea
|
selen |
2022-05-04 00:09:35 |
Docker |
dockerタグが付けられた新着投稿 - Qiita |
Rails+ReactのDocker環境(TechpitさんのUberEats風SPAアプリ) |
https://qiita.com/khinatas3/items/9cd426a2af3efc737d2b
|
docker |
2022-05-04 00:40:31 |
Docker |
dockerタグが付けられた新着投稿 - Qiita |
Docker ComposeでMySQL環境を構築する |
https://qiita.com/kaburankattara/items/7b5ec64d86f93b8db625
|
docker |
2022-05-04 00:17:09 |
Git |
Gitタグが付けられた新着投稿 - Qiita |
入社3年目、業務以外の知識がない自分がGWを捧げてサービス・アプリケーションを勉強してみた!〜4日目Git:リポジトリの作成からPull Requestまで〜 |
https://qiita.com/hugo-crt/items/24c4147d851cf30bcf22
|
pullrequest |
2022-05-04 00:52:03 |
Ruby |
Railsタグが付けられた新着投稿 - Qiita |
Rails+ReactのDocker環境(TechpitさんのUberEats風SPAアプリ) |
https://qiita.com/khinatas3/items/9cd426a2af3efc737d2b
|
docker |
2022-05-04 00:40:31 |
海外TECH |
Ars Technica |
Kindle e-readers finally (kind of) support ePub books |
https://arstechnica.com/?p=1851830
|
conversion |
2022-05-03 15:25:44 |
海外TECH |
Ars Technica |
Turning waste into gold drugs |
https://arstechnica.com/?p=1851681
|
reaction |
2022-05-03 15:13:35 |
海外TECH |
MakeUseOf |
Sticky Notifications: Get Virtual Stickies For Your Mac |
https://www.makeuseof.com/tag/sticky-notifications-stickies-for-mac/
|
quick |
2022-05-03 15:45:14 |
海外TECH |
MakeUseOf |
9 Online Tools to Compare Files and Find Differences |
https://www.makeuseof.com/online-tools-compare-files/
|
online |
2022-05-03 15:30:13 |
海外TECH |
MakeUseOf |
How to Remove the Drop Shadow From Desktop Icons in Windows 10 |
https://www.makeuseof.com/how-to-remove-drop-shadow-from-desktop-icons-windows-10/
|
windows |
2022-05-03 15:15:14 |
海外TECH |
DEV Community |
Using TensorFlowJS & HarperDB for Machine Learning |
https://dev.to/logicdudes/using-tensorflowjs-harperdb-for-machine-learning-1me1
|
Using TensorFlowJS amp HarperDB for Machine Learning Implementing a Dog Breed Classifier Using Stanford Dogs and MobileNet with HarperDB Custom Functions IntroHarperDB is an easy to use database solution that has a simple method of creating endpoints to interact with data called Custom Functions These Custom Functions can even be used to implement a machine learning algorithm to classify incoming data TensorFlowJS is a library released by Google that makes it possible to use JavaScript for machine learning so it can be done in the browser or on a NodeJS server like we ll be doing in this article Summary What We re Going To DoThis article will explain how to train and use a TensorFlowJS model to classify dog breeds with HarperDB Custom Functions using the Stanford Dogs dataset and MobileNetV as a base for transfer learning Stanford DogsThere s an awesome dataset that was released by Stanford with images of dogs The images are grouped into different folders each folder containing the name of the breed There are additional annotations available for bounding boxes as well but today we ll be focused solely on classifying the breed MobileNetThere s a SOTA state of the art model published by Google called MobileNet which is a relatively small model with the ability to classify over images It s built small so it ll run on mobile devices without taking up too many resources We ll be using version of this model which is available in the tensorflow models mobilenet package Transfer LearningTransfer learning is the technique of taking a pretrained model and training it to output new data Like teaching an old dog new tricks For that we ll be using tensorflow models knn classifier We ll be sending an image into MobileNet and getting out the logits which is the bit right before the classification Then we ll send those logits into a KNN Classifier which uses the K Nearest Neighbors algorithm to associate those logits with specific dog breeds Getting StartedIf that all sounds complicated don t worry This implementation will be quick and easy thanks to HarperDB Custom Functions Setup PrereqsA HarperDB AccountA HarperDB Local Database Clone the RepoClone this repo into your Custom Functions foldergit clone hdb src custom functions dogml Restart Custom FunctionsUse the link in the HarperDB Studio Functions page bottom left of the screen to refresh the projects Run setupThe training data and TensorFlowJS modules need to be installed This can be done via the setup endpoint If you go to http localhost dogml setup it ll start the setup You can check on the progress in the logs either in stdout from the locally running database or in the logs section of the Status page inside of the Studio The expected output of starting setup is success true message ML Setup Started This will use the HOME dogml directory in relation to the database for all of the training materials Be sure to wait for the ML Setup Complete note in the database logs Activate Run trainTo train the model visit the train endpoint by going to http localhost dogml train This will begin the model training You can see the status inside of the console logs similar to viewing the info during setup or inside of the logs table inside of the schema Verify ModelOnce the logs indicate that the training is complete you should be able to see the model appear in the models table in the schema Classify a Dog Breed Travel to the UI at http localhost dogml ui and try uploading an image of a dog one of the images in the HOME dogml training data Images directory will do The results should appear in the UI as well as in the classifications table Go Deeper Add New Training DataYou can add more training data by adding new images to the HOME dogml training data Images directory either by putting the image in the correct folder or making a new folder if it s a breed without a folder already present All images should be JPEGs Removing Training DataYou can also remove training data in the HOME dogml training data Images directory to better target specific breeds Update the ModelIf you modify the training data and use the train endpoint to create a new model be sure to then call the update endpoint at http localhost dogml update to ensure the new model is loaded into the classifier Train w GPUTo train the model faster use the train gpu endpoint at http localhost dogml train gpu This will take advantage of a CUDA Enabled Nvidia GPU to process the training mathematics quicker Be sure the necessary drivers and CUDA libraries are installedHere s a guide to install CUDA on Ubuntu ReviewThere you have it you ve just trained a machine learning model on dog breed data and can now use it to classify images of dogs and determine the breed To do this we used a HarperDB Custom Function and TensorFlowJS to train a MobileNet model on the Stanford Dogs dataset |
2022-05-03 15:33:46 |
海外TECH |
DEV Community |
[Conceito] - Sofra com Embasamento Matemático |
https://dev.to/zanfranceschi/conceito-sofra-com-embasamento-matematico-176o
|
Conceito Sofra com Embasamento MatemáticoConteúdo original nessa thread do TwitterEi dev Vou te mostrar como sofrer com embasamento na teoria de redes em eventos sociais pro caso de vocêser uma pessoa super introvertida assim como eu Émuito mais bonito e sexy sofrer matematicamente Geralmente a gente que éintrovertido sofre antecipadamente dependendo da possibilidade de interações entre pessoas no evento Ou seja quanto mais pessoas a ansiedade préevento cresce aritmeticamente A fórmula é I P x P ÷ I Interações e P pessoas Mas a gente sabe que o nosso sofrimento realizado quase nunca étão grande quanto o antecipado né O sofrimento realizado éo número real de interações durante o evento Nesse exemplo sóhouve interações de fato Sabendo do sofrimento antecipado e sofrimento realizado vocêconsegue obter uma razão que se chama densidade da rede que nada mais édo que o realizado ÷antecipado No caso seria ÷ que resulta na dízima periódica ou também Ou seja nesse exemplo vocêentendeu que pra calcular seu sofrimento énecessário obter o sofrimento antecipado ーque éo número possível de interações ーe do sofrimento realizado ーque éo número praticado de interações Faz um exercício de sofrimento agora calcula o quanto de sofrimento haveria num evento com pessoas e com interações realizadas Se vocêchegou atéaqui faz uma oração pra mim por favor Semana que vem vou participar dum evento em que não conheço ninguém E o pior de tudo éque nem sofrer direito antecipadamente eu posso pq não sei quantas pessoas irão 🥲 |
2022-05-03 15:32:33 |
Apple |
AppleInsider - Frontpage News |
How to use Dropbox to replace iCloud |
https://appleinsider.com/inside/icloud/tips/how-to-use-dropbox-to-replace-icloud?utm_medium=rss
|
How to use Dropbox to replace iCloudApple makes iCloud compelling but there are reasons to opt for Dropbox instead once you know how to use it fully Here s how to shift over You re an Apple user so you ve got an iCloud account even if it s just Apple s minimum GB version Apple leans on iCloud for syncing everything from app data and Apple Watch backups to SMS messages and Ringtones and it s generally very good for this kind of automatic sync However when it s documents you want synchronized and perhaps available not only to you but also to other people there are strong reasons to prefer Dropbox instead Read more |
2022-05-03 15:49:44 |
Apple |
AppleInsider - Frontpage News |
YouTuber given rare tour of Apple Fitness+ studios |
https://appleinsider.com/articles/22/05/03/youtuber-given-rare-tour-of-apple-fitness-studios?utm_medium=rss
|
YouTuber given rare tour of Apple Fitness studiosApple has given popular YouTuber iJustine a tour of its Apple Fitness studio in Santa Monica as well as interviews with the company s fitness executive and coaches Apple Fitness tourThe iPhone maker produces all of its weekly Apple Fitness content at a large studio in Santa Monica As shown in iJustine s video the studio is highly mechanized ーand trainers are essentially able to film their classes without a large technical team Read more |
2022-05-03 15:26:06 |
Apple |
AppleInsider - Frontpage News |
Apple Fitness+ executives talk about the service's ambition, and post-pandemic future |
https://appleinsider.com/articles/22/05/03/apple-fitness-executives-talk-about-the-services-ambition-and-post-pandemic-future?utm_medium=rss
|
Apple Fitness executives talk about the service x s ambition and post pandemic futureApple Fitness launched as coronavirus was keeping people from gyms but Apple says that the program is a moral responsibility that is only getting started It was September when Apple introduced the Apple Fitness exercise and workout app that ties in to a user s Apple Watch and AppleInsider would soon describe it as a godsend in those COVID days But the service was not rushed out to because of the lockdowns and Apple is not cutting back on it now that the pandemic is over The British GQ magazine recently toured the service s studio in Santa Monica Alongside hearing from Apple executives about its genesis the publication also learned that every trainer learned American Sign Language when they joined Read more |
2022-05-03 15:11:14 |
Apple |
AppleInsider - Frontpage News |
Activity Monitor in macOS is wrong about energy usage of Apple Silicon |
https://appleinsider.com/articles/22/05/03/activity-monitor-in-macos-is-wrong-about-energy-usage-of-apple-silicon?utm_medium=rss
|
Activity Monitor in macOS is wrong about energy usage of Apple SiliconThe Activity Monitor in macOS may not be as accurate with the data it provides to Apple Silicon users with a report claiming the tool cannot properly tell the difference between performance and efficiency cores The Activity Monitor offers a way for users and developers to tell what apps are using up the most resources and the most energy when performing tasks In testing of elements of the tool s features on a Mac running Apple Silicon it appears that a small mistake in core recognition may have wildly thrown off some of the results Using the Activity Monitor s CPU and Energy figures a user can potentially see that an app s code running just on efficiency cores are reportedly using more energy than performance cores when completing a task As efficiency cores are intended to be slower but with lower power usage than performance cores the result is fairly contradictory Read more |
2022-05-03 15:01:45 |
海外TECH |
Engadget |
SoundCloud buys an AI music company to help discover hidden gems |
https://www.engadget.com/soundcloud-buys-musiio-ai-music-curation-154640970.html?src=rss
|
SoundCloud buys an AI music company to help discover hidden gemsSoundCloud wants to make it easier to find must listen tunes buried deep in its catalog TechCrunchnotes the company has bought Musiio an AI music curation company The firm s technology uses AI to quot listen quot to songs tag them and slot them into playlists Ideally this improves the chances of discovering tracks you like whether they re from an established artist or a new bedroom DJ The companies didn t disclose the terms of the deal but said Musiio would quot become core quot to SoundCloud s discovery system Musiio will still offer its technology to other companies The two haven t revealed their exact plans or an integration timeline However it s easy to see advantages for both SoundCloud and listeners The company could use Musiio s AI to spot a breakthrough artist before a label or rival service poaches them It might also increase play counts for many songs and convince indie performers to stick around You meanwhile might listen to more varied playlists and discover a breakthrough song before it reaches the charts |
2022-05-03 15:46:40 |
海外TECH |
Engadget |
Spotify's podcast tech chief is leaving the company |
https://www.engadget.com/spotify-michael-mignano-anchor-podcast-153800296.html?src=rss
|
Spotify x s podcast tech chief is leaving the companyAnchor co founder Michael Mignano is leaving Spotify where he has led the podcast tech team over the last few years The company told The Verge nbsp that Mignano has submitted his resignation and will move on from Spotify at the end of June Spotify bought Anchor in for a reported million to help people create record and manage their own shows Mignano who co founded that company in started overseeing podcast tech at Spotify as a result He played a role in Spotify becoming a major player in podcasting and was involved in the company s push into live audio Around percent of new podcasts on Spotify are uploaded from Anchor The platform reportedly hosts million podcasts four times the number it had in Anchor now even supports video podcasts nbsp Although Spotify is firmly entrenched as a leader in the podcasting space that side of the business is going through a shakeup It was reported last month that Courtney Holt ーwho was involved in the Anchor acquisition and helped sign the Obamas Joe Rogan and Prince Harry and Meghan Markle to exclusive podcast deals ーhas stepped down as a Spotify executive but will remain as an advisor Lydia Polgreen also left her position as managing director of Gimlet Media in April Barack and Michelle Obama are moving on from Spotify as well The company says it opted not to offer them a new deal Their production company Higher Ground is looking for a new partner and is said to have been in discussions with the likes of Audible and iHeartMedia |
2022-05-03 15:38:00 |
Cisco |
Cisco Blog |
Unified Hybrid Cloud Deployments |
https://blogs.cisco.com/partner/unified-hybrid-cloud-deployments
|
Unified Hybrid Cloud DeploymentsAs customers migrate their applications to hybrid cloud environments they have great flexibility in choosing data centers to run their various workloads However with this flexibility comes enormous operational complexity as operations teams try to avoid costly downtime meet business SLAs and cost effectively scale underlying infrastructure distributed among multiple data centers to meet dynamic business needs |
2022-05-03 15:00:59 |
海外科学 |
NYT > Science |
Rocket Lab Grabs Booster Falling From Space With a Helicopter |
https://www.nytimes.com/2022/05/02/science/rocket-lab-launch-helicopter.html
|
Rocket Lab Grabs Booster Falling From Space With a HelicopterThe company aims to join Elon Musk s SpaceX in reusing rocket boosters which can lower costs and increase the frequency of launching to orbit |
2022-05-03 15:42:06 |
海外科学 |
NYT > Science |
Colorado River Reservoirs Are So Low, Government Is Delaying Releases |
https://www.nytimes.com/2022/05/03/climate/lake-powell-mead-water-drought.html
|
Colorado River Reservoirs Are So Low Government Is Delaying ReleasesThe anticipated decision will keep more water in Lake Powell on the Arizona Utah border instead of releasing it downstream to Lake Mead Both reservoirs are at their lowest points |
2022-05-03 15:05:24 |
ニュース |
BBC News - Home |
Energy firms face deadline in review of direct debit hikes |
https://www.bbc.co.uk/news/business-61309576?at_medium=RSS&at_campaign=KARANGA
|
debits |
2022-05-03 15:53:48 |
ニュース |
BBC News - Home |
Boris Johnson says there is more to do to help ease living costs |
https://www.bbc.co.uk/news/uk-politics-61306589?at_medium=RSS&at_campaign=KARANGA
|
johnson |
2022-05-03 15:07:47 |
ニュース |
BBC News - Home |
UK ambassador: 'Hard to be critical' of Brits in Ukraine |
https://www.bbc.co.uk/news/uk-61310816?at_medium=RSS&at_campaign=KARANGA
|
existential |
2022-05-03 15:40:25 |
ニュース |
BBC News - Home |
Nottinghamshire police boss Caroline Henry caught speeding five times |
https://www.bbc.co.uk/news/uk-england-nottinghamshire-61307006?at_medium=RSS&at_campaign=KARANGA
|
caroline |
2022-05-03 15:09:20 |
ニュース |
BBC News - Home |
WG Grace: Cricket legend has 10 matches wiped from iconic first-class record |
https://www.bbc.co.uk/sport/cricket/61308258?at_medium=RSS&at_campaign=KARANGA
|
WG Grace Cricket legend has matches wiped from iconic first class recordMore than a years after his death legendary cricketer WG Grace has runs wickets and two centuries wiped from the records books |
2022-05-03 15:08:30 |
北海道 |
北海道新聞 |
桂田社長、安全規程違反認める 家族向け文書で、定点連絡も怠る 知床・観光船事故 |
https://www.hokkaido-np.co.jp/article/676955/
|
知床半島 |
2022-05-04 00:28:03 |
北海道 |
北海道新聞 |
米バイオジェンCEO退任 アルツハイマー新薬難航 |
https://www.hokkaido-np.co.jp/article/676973/
|
難航 |
2022-05-04 00:01:26 |
北海道 |
北海道新聞 |
米3月求人が過去最高に 1154万件、人手不足で |
https://www.hokkaido-np.co.jp/article/676976/
|
人手不足 |
2022-05-04 00:14:00 |
北海道 |
北海道新聞 |
対中関税可否で意見募集 米、政権内に見直し論も |
https://www.hokkaido-np.co.jp/article/676977/
|
意見募集 |
2022-05-04 00:14:00 |
コメント
コメントを投稿