The document discusses using gRPC and Protocol Buffers to build fast and reliable APIs, describing how gRPC uses Protocol Buffers to define service interfaces and handle serialization, and allows building clients and servers in various languages that can communicate over the network through language-independent services. It provides examples of using gRPC to define and call both unary and streamin
機械学習とif文が地続きであることを解説しました。 ver.2 質問への回答を追加し、顧客価値の小問に図を追加してわかりやすくかみ砕きました。Read less
This document discusses using Ruby for distributed storage systems. It describes components like Bigdam, which is Treasure Data's new data ingestion pipeline. Bigdam uses microservices and a distributed key-value store called Bigdam-pool to buffer data. The document discusses designing and testing Bigdam using mocking, interfaces, and integration tests in Ruby. It also explores porting Bigdam-pool
脳型計算機雑談会での資料です 1. 大きなNNの学習はなぜ一様に成功するか 2. 敵対的生成ネットワーク(GAN)の解析 3. seq2seqによる可変長情報の埋め込み 4. Ladder Networkの解析 Read less
2012年の画像認識コンペティションILSVRCにおけるAlexNetの登場以降,画像認識においては畳み込みニューラルネットワーク (CNN) を用いることがデファクトスタンダードとなった.CNNは画像分類だけではなく,セグメンテーションや物体検出など様々なタスクを解くためのベースネットワークとしても広く利用されてきている.本講演では,AlexNet以降の代表的なCNNの変遷を振り返るとともに,近年提案されている様々なCNNの改良手法についてサーベイを行い,それらを幾つかのアプローチに分類し,解説する.更に,実用上重要な高速化手法について、畳み込みの分解や枝刈り等の分類を行い,それぞれ解説を行う. Recent Advances in Convolutional Neural Networks and Accelerating DNNs 第21回ステアラボ人工知能セミナー講演資料 http
The document summarizes recent research related to "theory of mind" in multi-agent reinforcement learning. It discusses three papers that propose methods for agents to infer the intentions of other agents by applying concepts from theory of mind: 1. The papers propose that in multi-agent reinforcement learning, being able to understand the intentions of other agents could help with cooperation and
Callbacks, Promises, and Coroutines (oh my!): Asynchronous Programming Patterns in JavaScript This talk takes a deep dive into asynchronous programming patterns and practices, with an emphasis on the promise pattern. We go through the basics of the event loop, highlighting the drawbacks of asynchronous programming in a naive callback style. Fortunately, we can use the magic of promises to escape f
3. WebPエンコードのパラメータ • デフォルト • PICTURE – Target_size = 0 – Sns_strength = 80 – Target_PSNR = 0. – Fiter_sharpness = 4 – Method = 4 Method = 4 – Filter strength = 35 Filter strength = 35 – Sns_strength = 50 • PHOTO – Filter_strength = 20 – Sns_strength = 80 – Filter_sharpness = 0 _ p – Fiter sharpness = 3 Fiter_sharpness 3 – Filter_type 0 – Filter strength = 30 – Partitions = 0 • DRAWING – Segments =
(slides from the O'Reilly webcast, see recording here: http://www.oreilly.com/pub/e/3425) The web is becoming increasingly image rich. Between high-resolution mobile screens, Pinterest-style design and big background graphics, the average image payload has more than doubled in the last three years. While visually appealing, these images carry a substantial performance cost, and — if not optimized
This document summarizes a microservices meetup hosted by @mosa_siru. Key points include: 1. @mosa_siru is an engineer at DeNA and CTO of Gunosy. 2. The meetup covered Gunosy's architecture with over 45 GitHub repositories, 30 stacks, 10 Go APIs, and 10 Python batch processes using AWS services like Kinesis, Lambda, SQS and API Gateway. 3. Challenges discussed were managing 30 microservices, ensur
[第3回分析コンペLT会 、オンライン開催] (https://kaggle-friends.connpass.com/event/220927/) での発表資料です。 画像コンペに出るうえで便利過ぎる timm(pytorch image models) の紹介をしました。
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
処理を実行中です
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