Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
AI 人工智慧學校分享
Search
blue chen
April 20, 2018
0
400
AI 人工智慧學校分享
blue chen
April 20, 2018
Tweet
Share
More Decks by blue chen
See All by blue chen
RelaJet Caption
iamblue
0
110
The future of hearing device - 聽覺領域新的機會及挑戰
iamblue
2
250
20180918-Digitime 導入MCU設計 催熟智慧語音辨識應用市場
iamblue
0
230
Pixnet hackthon - workshop
iamblue
0
110
Javascript -Full stack 物聯網開發
iamblue
0
130
MCS Lite 私有雲物聯網開發
iamblue
0
300
Blockchain for IoT 應用
iamblue
1
220
2017.01.16 Embedded system
iamblue
0
110
microlattice.js ( World of tech 2016 )
iamblue
1
290
Featured
See All Featured
Six Lessons from altMBA
skipperchong
29
4.1k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
34
2.5k
The Web Performance Landscape in 2024 [PerfNow 2024]
tammyeverts
10
920
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
508
140k
Navigating Team Friction
lara
190
15k
Intergalactic Javascript Robots from Outer Space
tanoku
273
27k
Chrome DevTools: State of the Union 2024 - Debugging React & Beyond
addyosmani
9
960
Unsuck your backbone
ammeep
671
58k
Keith and Marios Guide to Fast Websites
keithpitt
413
23k
Statistics for Hackers
jakevdp
799
220k
Agile that works and the tools we love
rasmusluckow
331
21k
For a Future-Friendly Web
brad_frost
180
10k
Transcript
洞洞 ⾒見見 未 來來 Deep learning on MicroControllerUnit Blue Chen
(
[email protected]
) 2018/04/20 ⼈人⼯工智慧學校
NN Layer Output Input 算法上的幾個關鍵 轉換 轉換 1. FFT (audio)
2. MFCC (audio) 3. RGB resize (vision) 4. Object detection (vision) 5. Face detection (vision) 1. IFFT (audio) 2. IMFCC (audio) 1. Model Size 2. Memory size 3. int / float (HW 加速)
NN Layer Output Input 以視覺辨識為例例 轉換 轉換 最後輸出 N x
1 維度
以聲⾳音輸出為例例 NN Layer Output Input 轉換 轉換
NN Layer Output Input 軟體 Support Effort 很龐⼤大 (先不論 AI
model 設計) 轉換 轉換 1.OS driver porting 2. 轉換器算法 3. openCV/ open VX 4. Dlib 5. FFmpeg (streaming) Coretex A - Android NN Coretex M - CMSIS NN 1. IFFT, IMFCC… 2. Trigger Application 極度不成熟 缺少比較新的 layer support Porting 底層⼈人才稀少 後續應⽤用層 ⼜又是另外⼀一層 Know how
不要隨便便說⾃自⼰己是 AI 公司 真正的 AI 公司,本質是超強的軟體團隊
Training Deployment RelaJet Zoo (各種模型) RelaJet 團隊開發架構 (視覺為例例) Caffe OpenCV
Linux OS Win/Mac Android OS IOS Tensorflow Theano Keras Caffe2 Caffe Pytorch … etc Base on Cortex A / X86 powerful platform OpenVX Base on CortexM 單⼼心片架構系統 語⾔言層 C / C++ C / C++ python/Js/ .. etc C / C++
One more thing….
RelaJet C series https://www.relajet.com/c-series
We provide 1. Vision tool 2. Tensorflow / Caffe /
Caffe2 / Keras model transfer tool 3. Capture Image and video tool 4. Training model tool RelaJet Omnism 全知 platform USD 79
Cifar10
聽覺產品敬請期待! Mail:
[email protected]
FB/ Line / Wechat ID: iambluechen
13 www.relajet.com