Tags » SaaS

"[Convnets, Keras] Training loss is stuck at the initial value when I start with a higher learning rate and slowly decrease it down to a learning rate for which the training loss decreases - Why?"

Posted By qwertz_guy. Detail: Something that I have observed multiple times when training neural networks (in particular convnets) is the following: Let’s say I have a small learning rate of 0.00001 and if I start with this LR then I can see how my loss is (starting from like 0.95) decreasing after each epoch to a very good value (<0.1) after let's say 20 epochs. 187 more words

Machine Learning

クラウドが導く“Beyond Finance”――、日本オラクル、SaaSによる予算管理サービスを提供へ [ #cloud ]

日本オラクル株式会社は27日、全社的な予算管理ソリューションをSaaSとして提供する「Oracle Enterprise Planning and Budgeting Cloud Service(Oracle EPBCS)」を発表した。すでにグローバルで多くの実績をもつ旧Hyperionをベースにしたクラウドソリューション「Oracle Planning and Budgeting Cloud Service(Oracle PBCS)」を拡張したサービスで、各部門ごとの円滑な予算策定を支援しつつ、全社共通の基盤を利用することにより、各部門の事業判断と組織全体の最適化を同時に実現することを目指す。

http://cloud.watch.impress.co.jp/docs/news/1012585.html

Topic

"Nvidia 1080, which one?"

Posted By DancesWithPizzas. Detail: So I’ve convinced myself that it’s time to upgrade my GPU for some ML fun. I’ve also convinced myself that I want a nvidia 1080. 120 more words

Machine Learning

"Keras - Image Augmentation"

Posted By Nader_Nazemi. Detail: Can someone please help me out figure what is the flow of applying “datagen = ImageDataGenerator()”For example I have: x_train , x_test , y_train , y_testHow do I augment x_train AND x_test using “datagen” ?Do I do the following:datagen = ImageDataGenerator(…stuff…) datagen.fit(x_train) datagen.fit(x_test)Then I create my Neural Network Model:model = Sequential() . 42 more words

Machine Learning

クラウドが導く“Beyond Finance”――、日本オラクル、SaaSによる予算管理サービスを提供へ [ #cloud ]

日本オラクル株式会社は27日、全社的な予算管理ソリューションをSaaSとして提供する「Oracle Enterprise Planning and Budgeting Cloud Service(Oracle EPBCS)」を発表した。すでにグローバルで多くの実績をもつ旧Hyperionをベースにしたクラウドソリューション「Oracle Planning and Budgeting Cloud Service(Oracle PBCS)」を拡張したサービスで、各部門ごとの円滑な予算策定を支援しつつ、全社共通の基盤を利用することにより、各部門の事業判断と組織全体の最適化を同時に実現することを目指す。

http://cloud.watch.impress.co.jp/docs/news/1012585.html

Service

"Keras - Image Augmentation"

Posted By Nader_Nazemi. Detail: Can someone please help me out figure what is the flow of applying “datagen = ImageDataGenerator()”For example I have: x_train , x_test , y_train , y_testHow do I augment x_train AND x_test using “datagen” ?Do I do the following:datagen = ImageDataGenerator(…stuff…) datagen.fit(x_train) datagen.fit(x_test)Then I create my Neural Network Model:model = Sequential() . 42 more words

Machine Learning