『您的瀏覽器不支援JavaScript功能,若網頁功能無法正常使用時,請開啟瀏覽器JavaScript狀態』

跳到主要內容區塊

工業技術研究院

:::

技術名稱: 深度學習訓練系統技術

技術簡介

工研院研發的「深度學習訓練系統」讓使用者在X86機器上可進行深度學習訓練,達成dataset management、neural network management、DNN training monitoring 等工作。深度學習的訓練仰賴高度平行運算來處理大量數據資料的訓練,除了利用GPU的高效能多核心的運算處理,結合「深度學習訓練系統」所搭載的深度學習框架(DNN Frameworks)以及提升效能訓練的進階功能,像是深度學習超參數自動調整、儲存設備與記憶體之間高速資料流動等技術,提供深度學習模型開發者一個方便且有效率的深度學習訓練環境,縮短訓練時程同時依舊維持高準確度。

Abstract

ITRI DNN Training System, IDTS, is a product of an integrated deep learning model training appliance, including the full deep learning training system software stack and the deep learning training optimization techniques on the commodity X86 GPU compatible machines and the DNN training GPUs. IDTS provides DNN developers an easy-to-use environment to do dataset management, training jobs management, and system environment monitoring. The developers can easily access the latest models and datasets, create, modify, visualize, and analyze complex DNNs, and train DNNs from TensorFlow for various topics, like image classification and object detection.

技術規格

1. 深度學習訓練系統包含OS、GPU driver、CUDA、cuDNN、cuBLAS、DNN frameworks(Caffe、Caffe2、TensorFlow)、DNN IDE 2. 計算資源支援NVIDIA GPU 3. DNN IDE:對neural network進行編輯和debug、監控DNN training過程;訓練完成後可進行模型壓縮,以利部署至不同計算資源的環境

Technical Specification

1. DNN Training System Software Stack for NVIDIA GPU 2. DNN IDE: Monitor, Analysis and Debug, Model Compression

技術特色

使用者可以透過DNN Training Appliance的系統軟體堆疊,縮短深度學習訓練時間、彈性並擴充使用多元深度學習資源、監控深度學習訓練過程並進行分析及修正,藉此改善使用者所需要的深度學習neural netwok。

應用範圍

AI應用開發商、AI應用服務商、伺服器硬體供應商、電信營運商、對AI應用有需求的企業機構(如:智慧醫療/智慧工廠)等。

接受技術者具備基礎建議(設備)

可營運管理GPU伺服器的資料中心、深度學習訓練開發基礎設備環境。

接受技術者具備基礎建議(專業)

GPU伺服器資料中心營運管理技術、深度學習訓練開發技術。

技術分類 資訊

聯絡資訊

聯絡人:李幸華(A80067) 企劃與推廣組

電話:16588 或 Email:evelyn_lee@itri.org.tw

客服專線:+886-800-45-8899

傳真:none

[{"text":"企業網","weight":13.0},{"text":"材化所","weight":11.5},{"text":"機械所","weight":10.0},{"text":"綠能所","weight":9.4},{"text":"生醫所","weight":8.0},{"text":"半導體","weight":6.2},{"text":"南分院","weight":5.0},{"text":"太陽能","weight":5.0},{"text":"課程","weight":5.0},{"text":"遠紅外線","weight":5.0},{"text":"雷射","weight":4.0},{"text":"LED","weight":4.0},{"text":"LED可見光","weight":3.0},{"text":"5G","weight":3.0},{"text":"工研人","weight":3.0},{"text":"電光所","weight":3.0},{"text":"綠能與環境研究所","weight":3.0},{"text":"機械","weight":3.0},{"text":"資通所","weight":2.0},{"text":"面板","weight":2.0},{"text":"文字轉語音","weight":2.0},{"text":"實習","weight":2.0},{"text":"無人機","weight":2.0},{"text":"生醫","weight":2.0},{"text":"3D","weight":2.0},{"text":"v2x","weight":2.0},{"text":"員工","weight":2.0},{"text":"地圖","weight":2.0},{"text":"太陽光電","weight":2.0},{"text":"材料與化工研究所","weight":1.0}]