Better performance with the tf.data API | TensorFlow Core
Overview GPUs and TPUs can radically reduce the time required to execute a single training step. Achieving peak performance requires an efficient input pipeline that delivers data for the next step before the current step has finished. The tf.data API helps to build flexible and efficient input pipelines. This document demonstrates how to use the tf.data API to build highly performant TensorFlow i
Speeding up Keras with tfrecord datasets
Alright. So you just got started with Keras with Tensorflow as a backend. Introducing GPU computing was quite simple so you started increasing the size of your datasets. Everything works fantastic, your GPU is happy and hungry for more, so you increase the dataset size even more to improve the robustness of your model. At a certain size, you hit the limit of your RAM and naturally you write a quic
Meet Horovod: Uber's Open Source Distributed Deep Learning Framework for TensorFlow
Data / ML, EngineeringMeet Horovod: Uber’s Open Source Distributed Deep Learning Framework for TensorFlowOctober 17, 2017 / Global Over the past few years, advances in deep learning have driven tremendous progress in image processing, speech recognition, and forecasting. At Uber, we apply deep learning across our business; from self-driving research to trip forecasting and fraud prevention, deep l
Raspberry Pi上でTensorFlowで簡単にディープラーニングを試せるツール「tensorflow-pi」でロボホンの幻の顔認識機能に挑戦してみた - karaage. [からあげ]
Raspberry PiとTensorFlowを使ったディープラーニング開発環境 以下の記事でRaspberry PiとTensorFlowを使ったディープラーニングの開発環境の構築方法を紹介しました。 記事の最後の方に、自前データの学習から、学習したニューラルネットワークで判別まで実施できる自前のパッケージ「tensorflow-pi」を紹介させていただきました。 ただ、READMEだけ見ても、意味不明な英語で良くわからないと思いますので、今回は実例を交えながらこのソフトの使い方を紹介しようと思います。 例題ですが、以前ネットで話題になった「ロボホン」の幻の顔認識機能があります。詳しくは以下参照下さい。 開発決定！というニュースは話題になったものの、その後続報がとんと聞かれません。そもそも技術的に不可能なんじゃという噂も聞こえてきたりしました。 奇しくもロボホンとRaspberry Pi
![Raspberry Pi上でTensorFlowで簡単にディープラーニングを試せるツール「tensorflow-pi」でロボホンの幻の顔認識機能に挑戦してみた - karaage. [からあげ]](https://cdn-ak-scissors.b.st-hatena.com/image/square/10adf88ff1495f3fbce26be21cda9215a10d1a47/height=288;version=1;width=512/https%3A%2F%2Fcdn-ak.f.st-hatena.com%2Fimages%2Ffotolife%2Fk%2Fkaraage%2F20170726%2F20170726015040.jpg)
Benchmarking CNTK on Keras: is it Better at Deep Learning than TensorFlow?
Benchmarking CNTK on Keras: is it Better at Deep Learning than TensorFlow? June 12, 2017 - Keras is a high-level open-source framework for deep learning, maintained by François Chollet, that abstracts the massive amounts of configuration and matrix algebra needed to build production-quality deep learning models. The Keras API abstracts a lower-level deep learning framework like Theano or Google’s
XLA: TensorFlow, Compiled! (TensorFlow Dev Summit 2017)
Speed is everything for effective machine learning, and XLA was developed to reduce training and inference time. In this talk, Chris Leary and Todd Wang describe how TensorFlow can make use of XLA, JIT, AOT, and other compilation techniques to minimize execution time and maximize computing resources. Visit the TensorFlow website for all session recordings: https://goo.gl/bsYmza Subscribe to the
1
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
Training on datasets too big to fit in RAM