Posts

Google Coral Edge TPU 工具包通過代理商Gravitylink向全球銷售!

Image
Coral是Google今年3月推出的AI智能硬件開發平臺,提供完整的本地 AI 工具包,可以輕鬆將用戶的想法從原型發展到生產:包含硬件組件、軟件工具和內容,從而幫助用戶在設備上本地創建、訓練和運行神經網絡。Coral 產品包括:全新 Coral Dev Board(單板計算機)、Coral USB Accelerator(USB 加速器)、Coral Camera(相機)已經在Google Coral Distributor 合作伙伴——Gravitylink的線上商城( https://store.gravitylink.com/global )向全球銷售。價格分別為:$149.99、$74.99、$24.99。 “初露鋒芒”的 Edge TPU 究竟是什麼? 縱觀近年來國內外在 AIoT 領域的實踐,可以看出隨著 AI 算力在邊緣設備上逐漸增強,過去跑在雲端服務器的機器學習能力逐漸下沈到邊緣設備上(如手機、汽車等移動設備),這種現象逐漸成為 AIoT 的一種流行趨勢。Edge TPU 可以說是 Google 專為邊緣計算推出的一款 ASIC 芯片。 Edge TPU 尺寸約為1美分硬幣的1/8大小,它可以在較小的物理尺寸及功耗範圍內提供強勁的性能。其最大優勢,在於搭載該芯片的任意一款產品均可本地加速設備機器學習推理,這樣不僅可以保護隱私,消除對持久網絡連接的需要,更能減少延遲和耗能功率,最大限度發揮性能。 除此之外,內置 Edge TPU 的 Coral Dev Board 等產品完美支持 TensorFlow Lite 在設備上使用。作為專為移動和物聯網設備而設計的輕量級開源學習框架,TensorFlow Lite 目前已部署在全球超過20億臺設備上,越來越多原生 Google 應用和開發者團隊都在使用它。目的就是:減小 AI 模型的大小,在不同硬件中提高模型推理速度。 3款內置 Edge TPU 的在售硬件分別是哪些? Coral 在發布之初,發布了五種產品,幫助團隊從新產品開發到現有設計的增強,最終達到擴大生產規模。目前有三款產品已經可以購買: ·Coral Dev Board(單板計算機) Coral Dev Board 是一款搭載了 Edge TPU 的單板計算機,功能非常豐富。開發板分為 SOM 模塊和...

how to make your phone be smarter? with or without Coral Dev Board

Image
I'd like to introduce a project of my team—— it's an app called Model Play( available on Google Play), that can run machine learning model on mobile phones, just like object detection, image classification and more. it's very fun to experience machine learning just buy our smartphones. More about Model Play: Model Play is a global platform for discovering, sharing, and experiencing easy to use machine learning models. The app built-in with a fast, simple, and small control tool that works with AI hardware, like Google Coral Dev Board. It allows running machine learning models with real-time performance entirely in a smartphone. In Model Play, users could subscribe, download, and submit their trained ML models, set models as paid or free for other users to download and get earnings by the models. Model Play offers an open, free, sharing platform, connects you with millions of global AI developers. It inspires and helps developers bring AI into more applications from...

Where to Buy Google Coral Dev Board?

Image
Surely this new tech is too expensive for hobbyists? Surprisingly, no. The Coral Dev Board costs just $149.99, a rival in cost to similar high-end development boards. The USB Accelerator is cheaper still, making it a great alternative to those unwilling to buy a whole new development board. You can pick up the components from Google partner’s Gravitylink Store website: https://store.gravitylink.com/global Dev Board: $149.99 USB Accelerator: $74.99 Camera Module: $24.99 Google Coral Dev Board The Coral Dev Board is a single-board computer with a removable system-on-module (SOM) that contains eMMC, SOC, wireless radios, and Google’s Edge TPU. You can use the Dev Board as a single-board computer when you need accelerated ML processing in a small form factor, but it also serves as an evaluation kit for the SOM. You can use the Dev Board to prototype internet-of-things (IOT) devices and other embedded systems that demand fast on-device ML inferencing, and then scale to ...

Call For Machine Learning Models——Model Play

Image
submit your trained models and get Google coral dev board as reward! Model Play is a global platform for discovering, sharing, and experiencing easy to use machine learning models. It’s compatible with Google Coral Dev Board, with a fast, simple, and small control tool in Model Play app that works with AI hardware. In Model Play, developers can submit their own trained ML models, subscribe to and download models of their own interest, or retrain models with inspiring and helping developers bring AI into more applications from prototype to product. Model Play is now calling for ML models. Submit your ML models and earn revenue?per download on Model Play! Rewards: 1)PICO-PI-IMX7-START KIT($179): The top 20 participants who submitted the ML model(passed our verification) will get this development board*1; 2)Google Coral USB Accelerator($74.99): Submit 3 ML models (passed our verification) will get 1 Coral USB Accelerator; (Top 100 only) 3)Google Coral Dev Board($149.99): S...

submit your trained models and get Google coral dev board as reward!

yep, it's a reward competition, submit your trained models Model Play is now calling for ML models. Submit your ML models and earn revenue   per download on Model Play! Rewards: 1)PICO-PI-IMX7-START KIT($179): The top 20 participants who submitted the ML model(passed our verification) will get this development board*1; 2)Google Coral USB Accelerator($74.99): Submit 3 ML models (passed our verification) will get 1 Coral USB Accelerator; (Top 100 only) 3)Google Coral Dev Board($149.99): Submit 9 ML models (passed our verification) will get 1 Coral Dev Board; (Top 50 only) Developers around the world could submit trained ML models to Model Play. The models should be TensorFlow Lite models and can be compiled to run on the Google Coral Dev Board. learn more: https://model.gravitylink.com/contribution

Google Coral products are all available at Gravitylink Store!

Image
Good news! Google Coral products are all available at Gravitylink Store! Coral Dev Board, Coral USB Acceletator and Coral Camera could deliver to world wide!  Google AIY Kits are on hot sale too! Coral Dev Board A development board to quickly prototype on-device ML products. Scale from prototype to production with a removable system-on-module (SOM). TensorFlow Lite models can be compiled to run on the Coral Dev Board. You can use the Dev Board to prototype internet-of-things (IOT) devices and other embedded systems that demand fast on-device ML inferencing. Edge TPU Key Benefits: ·High Speed TensorFlow Lite Inferencing ·Low Power ·Small Footprint Scale From Prototype to Production Considers your manufacturing needs. The SOM can be removed from the baseboard, ordered in bulk, and integrated into your hardware. Includes a Full System SOC + ML + Connectivity all on the board running a derivative of Debian Linux we call Mendel, so you can run your favo...

Google I/O Inspiration專場談及的10條人生經驗

Image
今年的Google I/O大會不僅讓開發人員參與其中。 搖滾明星,宇航員和圖靈獎獲獎者參加了十幾場靈感·頭腦風暴,與會者了解了技術如何塑造未來,從音樂到藝術再到創意。 以下是從這些會談中學到的一些經驗: 創造力的關鍵是像孩子一樣思考。 獲得奧斯卡獎的動畫師格倫基恩創造了像Ariel,The Beast和Pocahontas這樣的心愛人物。 他告訴與會者,無論你的工作是什么,與你內心的孩子保持聯系很重要。 “我們都有這個六歲小孩,這與你今天和你在做什么有關,”他說。 “不要忘記你小時候冒險的那部分。” 如果技術是答案,那么問題是什么? 這就是藝術家和研究員Sougwen Chung工作背后的論點,他編寫了機器人與她合作創作藝術品。她與藝術工作室Onformative的聯合創始人兼創意負責人Cedric Kiefer以及Google Arts&Culture藝術家+機器智能計劃的聯合負責人Kenric McDowell一同發表了演講。三人談到了藝術家與人工智能之間的關系,以及人工智能能否完全取代藝術家。 “這就是你在實踐中如何在你的藝術中實際使用技術的問題,”塞德里克說。 “你只是寫了一點代碼,然后按'藝術,藝術,藝術,更多藝術嗎?'這不完全是這樣的。” 要大膽,敢于思考。 宇航員Mae Jemison是第一位進入太空的有色女性,她領導著100年的星艦,這項計劃旨在確保人類可以在下個世紀前往另一位明星。 “當你看太空探索時,它的大膽有所不同,”她說。 “我認為火星不會讓我們足夠努力。”她加入了哈佛天體物理學家Sheperd Doeleman的演講,他幫助建造了第一張黑洞照片 - 一個大膽的項目本身。 技術可以推動你創造新事物。 YACHT樂隊的歌手Claire Evans在Google的Magenta的幫助下將機器學習融入到新專輯的創作中,這是一個研究ML在藝術中的作用的研究項目。 她使用Magenta根據YACHT的后期目錄創作新的旋律。 “它強行將我們推到了我們的舒適區之外,迫使我們以不同的方式玩耍,并以不同的方式思考我們的工作方式,”她說。 加入Magenta的Google員工Adam Roberts和Jesse Engel,以及The Flaming Lips的Wayne Coyne,他討論了他的樂隊如何...