SlideShare uma empresa Scribd logo
1 de 44
Baixar para ler offline
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Masaki Samejima
Machine Learning Solutions Architect, Amazon Web Services Japan.
2019.2.14
Developers Summit 2019
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Agenda
•
•
•
•
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
Demographic Data
Facial Landmarks
Sentiment Expressed
Image Quality
General Attributes
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
2012
SuperVision[1]
ILSVRC2012
[1] A. Krizhevsky, et al., Imagenet classification with deep convolutional neural networks, NIPS 2012.
[2] R Girshick, et al., Rich feature hierarchies for accurate object detection and semantic segmentation, CVPR 2014.
[3] I.J. Goodfellow, et al., Generative Adversarial Nets, NIPS 2014.
[4] V. Badrinarayanan, et al, SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. PAMI 2017
2014
R-CNN[2] Pascal
VOC GAN[3] SegNet[4]
2015
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
https://gluon-cv.mxnet.io/model_zoo/classification.html
senet_154
resnet_v1d
resnet_v1c
resnet_v1b
resnet_v1
densenet
darknet
VGG
resnet_v2
mobilenet
mobilenetv2
0.80
0.75
0.70
Accuracy
1000 2000 #sample/sec.3000 4000
• ImageNet 80%
• V100 GPU
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
https://gluon-cv.mxnet.io/model_zoo/detection.html
mAP
10 100
#sample/sec.
40
35
30
yolo3
faster_rcnn
ssd
• (IoU )
mAP 30-40%
•
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
https://gluon-cv.mxnet.io/model_zoo/segmentation.html
0
10
20
30
40
50
60
70
80
90
100
fcn_resnet101 psp_resnet101 deeplab_resnet101 fcn_resnet101 psp_resnet101 deeplab_resnet101 deeplab_resnet152
COCO VOC
IoU
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
3 [1]
[1] B. Tekin, et al., Real-Time Seamless Single Shot 6D Object Pose Prediction, CVPR 2018.
[2] R. Girdhar, et al., Detect-and-Track: Efficient Pose Estimation in Videos, CVPR 2018.
[3] L. Chen, et al., MaskLab: Instance Segmentation by Refining Object Detection with
Semantic and Direction Features, CVPR 2018.
[2]
[3]
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
GANNoise
Text-to-image [3]
(and Image-to-text)[1]
[2]
[1] P. Isola, et al., Image-to-Image Translation with Conditional Adversarial Nets, CVPR 2017.
[2] C. Ledig, et al., Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network, CVPR 2017.
[3] S. Reed, et al., Generative Adversarial Text to Image Synthesis, ICML 2016.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Saliency ( ) [1]
[1] N. Liu, et al., PiCANet: Learning Pixel-wise Contextual Attention for Saliency Detection, CVPR 2018.
[2] Z. Li, et al., MegaDepth: Learning Single-View Depth Prediction from Internet Photos, CVPR 2018.
[2]
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7 8 9 1011121314151617181920
ID
[1] O. Vinyals, et al., Matching Networks for One Shot Learning, arXiv:1606.04080
•
• [1]
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
• Deep Learning
•
X. Yuan, et al., Adversarial Examples: Attacks and Defenses for Deep Learning, IEEE Trans Neural Netw Learn Syst. 2019.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
•
•
ONNX
AutoML
Define-by-run
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
TensorFlow models
TF slim
GluonCV ChainerCV PyTorchCV
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
ResNet (Gluon vs MXNet)
num_unit = len(units)
assert(num_unit == num_stages)
data = mx.sym.Variable(name='data')
if dtype == 'float32':
data = mx.sym.identity(data=data, name='id')
else:
if dtype == 'float16':
data = mx.sym.Cast(data=data, dtype=np.float16)
data = mx.sym.BatchNorm(data=data, fix_gamma=True, eps=2e-5, momentum=bn_mom, name='bn_data')
(nchannel, height, width) = image_shape
if height <= 32: # such as cifar10
body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(3, 3), stride=(1,1), pad=(1, 1),
no_bias=True, name="conv0", workspace=workspace)
else: # often expected to be 224 such as imagenet
body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(7, 7), stride=(2,2), pad=(3, 3),
no_bias=True, name="conv0", workspace=workspace)
body = mx.sym.BatchNorm(data=body, fix_gamma=False, eps=2e-5, momentum=bn_mom, name='bn0')
body = mx.sym.Activation(data=body, act_type='relu', name='relu0')
body = mx.sym.Pooling(data=body, kernel=(3, 3), stride=(2,2), pad=(1,1), pool_type='max')
for i in range(num_stages):
body = residual_unit(body, filter_list[i+1], (1 if i==0 else 2, 1 if i==0 else 2), False,
name='stage%d_unit%d' % (i + 1, 1), bottle_neck=bottle_neck, workspace=workspace,
memonger=memonger)
for j in range(units[i]-1):
body = residual_unit(body, filter_list[i+1], (1,1), True, name='stage%d_unit%d' % (i + 1, j + 2),
bottle_neck=bottle_neck, workspace=workspace, memonger=memonger)
bn1 = mx.sym.BatchNorm(data=body, fix_gamma=False, eps=2e-5, momentum=bn_mom, name='bn1')
relu1 = mx.sym.Activation(data=bn1, act_type='relu', name='relu1')
MXNet
from mxnet.gluon.model_zoo import vision
resnet18 = vision.resnet18_v1()
Gluon
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
ONNX (Open Neural Network Exchange)
MXNet
Caffe2
PyTorch
TF
CNTKCoreML
Tensor
RT
NGraph
SNPE
•
ONNX ONNX
•
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
ONNX
Protocol Buffers
•
•
• API
Protocol Buffers
Graph Operator Tensor, …
Operator Definitions
ONNX Python API
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Define-and-run Define-by-run
• Define-and-run
•
• TensorFlow, MXNet
• Define-by-run
•
• Chainer PyTorch, TensorFlow, MXNet
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Define-and-run Define-by-run
Define-and-run Define-by-run
def our_function(A, B):
C = A + B
return C
A = Load_Data_A()
B = Load_Data_B()
result = our_function(A, B)
A = placeholder()
B = placeholder()
C = A + B
our_function =
compile(inputs=[A, B], outputs =[C])
A = Load_Data_A()
B = Load_Data_B()
result = our_function(A, B)
https://gluon.mxnet.io/chapter07_distributed-learning/hybridize.html
Define
Run
Define, Run
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Define-by-run
Define-and-run Define-by-run
def our_function(A, B):
C = A + B
return C
A = Load_Data_A()
B = Load_Data_B()
result = our_function(A, B)
A = placeholder()
B = placeholder()
C = A + B
our_function =
compile(inputs=[A, B], outputs =[C])
A = Load_Data_A()
B = Load_Data_B()
result = our_function(A, B)
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
AutoML
•
• , etc.
D. Bayor, et al., TFX: A TensorFlow-Based Production-Scale Machine Learning Platform, KDD 2017.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
AutoML
• AutoML
• ICML 2014 AutoML *
•
•
• Meta-Learning, Learning to learn
* https://sites.google.com/site/automlwsicml14/
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
AutoML
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
AutoML Amazon Forecast
User
CSV file
1. S3
2. Forecast
3. Forecast
4.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
Model Server
Interpretable ML
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Model Server
•
•
Model Server
•
• REST/RPC
Model Server Mobile client
Deploy
REST/RPC
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
TensorFlow Serving
[1] C. Olston, et al., TensorFlow-Serving: Flexible, High-Performance ML Serving, NIPS 2017.
• Controller, Synchronizer Serving job
• Router Serving job
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
MXNet Model Server
https://aws.amazon.com/jp/blogs/news/model-server-for-apache-mxnet-v1-0-released/
•
REST
API
• MMS 1.0
1,000
MMS 1.0
MMS 0.4
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
•
• AWS, SageMaker Neo
• Nvidia, TensorRT
Raspberry Pi
ResNet18 Mobilenet
11.5x
2.2x
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
SageMaker Neo / TVM
• Operator Fusion
• Data Layout Transformation
4x4 4x4
• Tensor Expression and Schedule Space
• Nested Parallelism with Cooperation
• etc…
T. Chen, et al., TVM: An Automated End-to-End Optimizing Compiler for Deep Learning, OSDI 2018.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
TensorRT
• Layer & Tensor Fusion
1
• FP16 and INT8 Precision Calibration
FP32 FP16 INT8
• Kernel Auto-Tuning
• Dynamic Tensor Memory
• Multi Stream Execution
https://devblogs.nvidia.com/tensorrt-3-faster-tensorflow-inference/
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Interpretable ML:
: SVM GBT
C. Molnar, Interpretable Machine Learning, https://christophm.github.io/interpretable-ml-book/
>900< 900
< 2000 km2 > 2000 km2
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Interpretable ML for computer vision
•
•
M.T. Ribeiro, et al., Anchors: High-Precision Model-Agnostic Explanations, AAAI 2018.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
•
• 1 1
•
• AWS Inferentia
• Intel Nervana
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Machine Learning on FPGA
• FPGA
• AWS F1 instance Amazon Machine
Image
•
Loop tiling [1]
[1] C. Zhang, et al., Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks, FPGA 2015.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
• GPU
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
•
•
AutoML AI
•
© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
https://amzn.to/aws_dev

Mais conteúdo relacionado

Semelhante a 【14-C-7】コンピュータビジョンを支える深層学習技術の新潮流

Semelhante a 【14-C-7】コンピュータビジョンを支える深層学習技術の新潮流 (20)

Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
 
MongoDB .local London 2019: Using AWS to Transform Customer Data in MongoDB i...
MongoDB .local London 2019: Using AWS to Transform Customer Data in MongoDB i...MongoDB .local London 2019: Using AWS to Transform Customer Data in MongoDB i...
MongoDB .local London 2019: Using AWS to Transform Customer Data in MongoDB i...
 
Amazon SageMaker Algorithms: Machine Learning Week San Francisco
Amazon SageMaker Algorithms: Machine Learning Week San FranciscoAmazon SageMaker Algorithms: Machine Learning Week San Francisco
Amazon SageMaker Algorithms: Machine Learning Week San Francisco
 
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
 
[NEW LAUNCH!] Introducing Amazon Elastic Inference: Reduce Deep Learning Infe...
[NEW LAUNCH!] Introducing Amazon Elastic Inference: Reduce Deep Learning Infe...[NEW LAUNCH!] Introducing Amazon Elastic Inference: Reduce Deep Learning Infe...
[NEW LAUNCH!] Introducing Amazon Elastic Inference: Reduce Deep Learning Infe...
 
Time series modeling workd AMLD 2018 Lausanne
Time series modeling workd AMLD 2018 LausanneTime series modeling workd AMLD 2018 Lausanne
Time series modeling workd AMLD 2018 Lausanne
 
Real Time and Offline Applications with GraphQL
Real Time and Offline Applications with GraphQLReal Time and Offline Applications with GraphQL
Real Time and Offline Applications with GraphQL
 
Deep Learning at AWS: Embedding & Attention Models
Deep Learning at AWS: Embedding & Attention ModelsDeep Learning at AWS: Embedding & Attention Models
Deep Learning at AWS: Embedding & Attention Models
 
AMF305_Autonomous Driving Algorithm Development on Amazon AI
AMF305_Autonomous Driving Algorithm Development on Amazon AIAMF305_Autonomous Driving Algorithm Development on Amazon AI
AMF305_Autonomous Driving Algorithm Development on Amazon AI
 
엔터프라이즈를 위한 머신러닝 그리고 AWS (김일호 솔루션즈 아키텍트, AWS) :: AWS Techforum 2018
엔터프라이즈를 위한 머신러닝 그리고 AWS (김일호 솔루션즈 아키텍트, AWS) :: AWS Techforum 2018엔터프라이즈를 위한 머신러닝 그리고 AWS (김일호 솔루션즈 아키텍트, AWS) :: AWS Techforum 2018
엔터프라이즈를 위한 머신러닝 그리고 AWS (김일호 솔루션즈 아키텍트, AWS) :: AWS Techforum 2018
 
Amazon SageMaker
Amazon SageMakerAmazon SageMaker
Amazon SageMaker
 
Enhanced Media Workflows Using Amazon AI
Enhanced Media Workflows Using Amazon AIEnhanced Media Workflows Using Amazon AI
Enhanced Media Workflows Using Amazon AI
 
Working with Amazon SageMaker Algorithms for Faster Model Training
Working with Amazon SageMaker Algorithms for Faster Model TrainingWorking with Amazon SageMaker Algorithms for Faster Model Training
Working with Amazon SageMaker Algorithms for Faster Model Training
 
ML Workflows with Amazon SageMaker and AWS Step Functions (API325) - AWS re:I...
ML Workflows with Amazon SageMaker and AWS Step Functions (API325) - AWS re:I...ML Workflows with Amazon SageMaker and AWS Step Functions (API325) - AWS re:I...
ML Workflows with Amazon SageMaker and AWS Step Functions (API325) - AWS re:I...
 
Advanced Patterns in Microservices Implementation with Amazon ECS - CON402 - ...
Advanced Patterns in Microservices Implementation with Amazon ECS - CON402 - ...Advanced Patterns in Microservices Implementation with Amazon ECS - CON402 - ...
Advanced Patterns in Microservices Implementation with Amazon ECS - CON402 - ...
 
Amazon SageMaker 기반 고품질 데이터 생성 및 심화 기계학습 기법 - 김필호 솔루션즈 아키텍트, AWS / 강정희 솔루션즈 아...
Amazon SageMaker 기반 고품질 데이터 생성 및 심화 기계학습 기법 - 김필호 솔루션즈 아키텍트, AWS / 강정희 솔루션즈 아...Amazon SageMaker 기반 고품질 데이터 생성 및 심화 기계학습 기법 - 김필호 솔루션즈 아키텍트, AWS / 강정희 솔루션즈 아...
Amazon SageMaker 기반 고품질 데이터 생성 및 심화 기계학습 기법 - 김필호 솔루션즈 아키텍트, AWS / 강정희 솔루션즈 아...
 
From notebook to production with Amazon Sagemaker
From notebook to production with Amazon SagemakerFrom notebook to production with Amazon Sagemaker
From notebook to production with Amazon Sagemaker
 
Amazon SageMaker 內建機器學習演算法 (Level 400)
Amazon SageMaker 內建機器學習演算法 (Level 400)Amazon SageMaker 內建機器學習演算法 (Level 400)
Amazon SageMaker 內建機器學習演算法 (Level 400)
 
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
 
Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Te...
Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Te...Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Te...
Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Te...
 

Mais de Developers Summit

【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~
【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~
【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~
Developers Summit
 

Mais de Developers Summit (20)

【18-A-2】ゲーミフィケーション・エバンジェリストが見る「あなたの技術力が“ワクワクするサービス”に変わる未来」
【18-A-2】ゲーミフィケーション・エバンジェリストが見る「あなたの技術力が“ワクワクするサービス”に変わる未来」【18-A-2】ゲーミフィケーション・エバンジェリストが見る「あなたの技術力が“ワクワクするサービス”に変わる未来」
【18-A-2】ゲーミフィケーション・エバンジェリストが見る「あなたの技術力が“ワクワクするサービス”に変わる未来」
 
【C-2・小林様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~
【C-2・小林様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~【C-2・小林様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~
【C-2・小林様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~
 
【C-2・醍醐様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~
【C-2・醍醐様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~【C-2・醍醐様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~
【C-2・醍醐様】AIとAPIがITインフラにもたらす変化 ~プログラマブルなクラウド型Wi-Fi~
 
【B-4】オープンソース開発で、フリー静的解析ツールを使ってみる
【B-4】オープンソース開発で、フリー静的解析ツールを使ってみる【B-4】オープンソース開発で、フリー静的解析ツールを使ってみる
【B-4】オープンソース開発で、フリー静的解析ツールを使ってみる
 
【B-6】Androidスマホの生体認証の脆弱性、調べてみたらよくある話だった。
【B-6】Androidスマホの生体認証の脆弱性、調べてみたらよくある話だった。【B-6】Androidスマホの生体認証の脆弱性、調べてみたらよくある話だった。
【B-6】Androidスマホの生体認証の脆弱性、調べてみたらよくある話だった。
 
【13-B-6】Hondaの生産技術屋さんがソフトウェア開発でアジャイルを初導入し組織変革に挑戦
【13-B-6】Hondaの生産技術屋さんがソフトウェア開発でアジャイルを初導入し組織変革に挑戦【13-B-6】Hondaの生産技術屋さんがソフトウェア開発でアジャイルを初導入し組織変革に挑戦
【13-B-6】Hondaの生産技術屋さんがソフトウェア開発でアジャイルを初導入し組織変革に挑戦
 
【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール
【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール
【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール
 
【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール
【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール
【15-E-7】セキュアな環境でDevOpsを実現する厳選ツール
 
【14-E-3】セキュリティ・テストの自動化によるDevSecOpsの実現 (デモ有)
【14-E-3】セキュリティ・テストの自動化によるDevSecOpsの実現 (デモ有)【14-E-3】セキュリティ・テストの自動化によるDevSecOpsの実現 (デモ有)
【14-E-3】セキュリティ・テストの自動化によるDevSecOpsの実現 (デモ有)
 
【15-D-2】デンソーのMaaS開発~アジャイル開発で顧客との協調・チームビルディング・実装概要~
【15-D-2】デンソーのMaaS開発~アジャイル開発で顧客との協調・チームビルディング・実装概要~【15-D-2】デンソーのMaaS開発~アジャイル開発で顧客との協調・チームビルディング・実装概要~
【15-D-2】デンソーのMaaS開発~アジャイル開発で顧客との協調・チームビルディング・実装概要~
 
【14-C-8】みんなの暮らしを支えるAmazon S3の裏側、お伝えします
【14-C-8】みんなの暮らしを支えるAmazon S3の裏側、お伝えします【14-C-8】みんなの暮らしを支えるAmazon S3の裏側、お伝えします
【14-C-8】みんなの暮らしを支えるAmazon S3の裏側、お伝えします
 
【15-B-7】無意味なアラートからの脱却 ~ Datadogを使ってモダンなモニタリングを始めよう ~
【15-B-7】無意味なアラートからの脱却 ~ Datadogを使ってモダンなモニタリングを始めよう ~【15-B-7】無意味なアラートからの脱却 ~ Datadogを使ってモダンなモニタリングを始めよう ~
【15-B-7】無意味なアラートからの脱却 ~ Datadogを使ってモダンなモニタリングを始めよう ~
 
【15-A-1】ドラゴンクエストXを支える失敗事例
【15-A-1】ドラゴンクエストXを支える失敗事例【15-A-1】ドラゴンクエストXを支える失敗事例
【15-A-1】ドラゴンクエストXを支える失敗事例
 
【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~
【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~
【15-A-5】ゲーミフィケーションエバンジェリストが説く、アプリ開発で見落としがちな「おもてなし」とは~面白さを伝える × 面白く魅せる~
 
【B-2】福岡発Node.jsで支える大規模システム!〜「誰ガ為のアルケミスト」と歩んだ三年〜
【B-2】福岡発Node.jsで支える大規模システム!〜「誰ガ為のアルケミスト」と歩んだ三年〜【B-2】福岡発Node.jsで支える大規模システム!〜「誰ガ為のアルケミスト」と歩んだ三年〜
【B-2】福岡発Node.jsで支える大規模システム!〜「誰ガ為のアルケミスト」と歩んだ三年〜
 
【B-5】モダンな開発を実現するツールチェーンのご紹介
【B-5】モダンな開発を実現するツールチェーンのご紹介【B-5】モダンな開発を実現するツールチェーンのご紹介
【B-5】モダンな開発を実現するツールチェーンのご紹介
 
【C-2】メモリも、僕のキャパシティも溢れっぱなし。。2年目エンジニアが実現した機械学習
【C-2】メモリも、僕のキャパシティも溢れっぱなし。。2年目エンジニアが実現した機械学習【C-2】メモリも、僕のキャパシティも溢れっぱなし。。2年目エンジニアが実現した機械学習
【C-2】メモリも、僕のキャパシティも溢れっぱなし。。2年目エンジニアが実現した機械学習
 
【A-2】とあるマーケティング部隊とデータエンジニアのデータドリブンへの道
【A-2】とあるマーケティング部隊とデータエンジニアのデータドリブンへの道【A-2】とあるマーケティング部隊とデータエンジニアのデータドリブンへの道
【A-2】とあるマーケティング部隊とデータエンジニアのデータドリブンへの道
 
【B-2】AI時代におけるエンジニアの生存戦略
【B-2】AI時代におけるエンジニアの生存戦略【B-2】AI時代におけるエンジニアの生存戦略
【B-2】AI時代におけるエンジニアの生存戦略
 
【B-2】AI時代のエンジニア生存戦略
【B-2】AI時代のエンジニア生存戦略【B-2】AI時代のエンジニア生存戦略
【B-2】AI時代のエンジニア生存戦略
 

Último

Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Safe Software
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
panagenda
 

Último (20)

Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontology
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering Developers
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 

【14-C-7】コンピュータビジョンを支える深層学習技術の新潮流

  • 1. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Masaki Samejima Machine Learning Solutions Architect, Amazon Web Services Japan. 2019.2.14 Developers Summit 2019
  • 2. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Agenda • • • •
  • 3. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 4. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • Demographic Data Facial Landmarks Sentiment Expressed Image Quality General Attributes
  • 5. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 2012 SuperVision[1] ILSVRC2012 [1] A. Krizhevsky, et al., Imagenet classification with deep convolutional neural networks, NIPS 2012. [2] R Girshick, et al., Rich feature hierarchies for accurate object detection and semantic segmentation, CVPR 2014. [3] I.J. Goodfellow, et al., Generative Adversarial Nets, NIPS 2014. [4] V. Badrinarayanan, et al, SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. PAMI 2017 2014 R-CNN[2] Pascal VOC GAN[3] SegNet[4] 2015
  • 6. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://gluon-cv.mxnet.io/model_zoo/classification.html senet_154 resnet_v1d resnet_v1c resnet_v1b resnet_v1 densenet darknet VGG resnet_v2 mobilenet mobilenetv2 0.80 0.75 0.70 Accuracy 1000 2000 #sample/sec.3000 4000 • ImageNet 80% • V100 GPU
  • 7. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://gluon-cv.mxnet.io/model_zoo/detection.html mAP 10 100 #sample/sec. 40 35 30 yolo3 faster_rcnn ssd • (IoU ) mAP 30-40% •
  • 8. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://gluon-cv.mxnet.io/model_zoo/segmentation.html 0 10 20 30 40 50 60 70 80 90 100 fcn_resnet101 psp_resnet101 deeplab_resnet101 fcn_resnet101 psp_resnet101 deeplab_resnet101 deeplab_resnet152 COCO VOC IoU
  • 9. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 3 [1] [1] B. Tekin, et al., Real-Time Seamless Single Shot 6D Object Pose Prediction, CVPR 2018. [2] R. Girdhar, et al., Detect-and-Track: Efficient Pose Estimation in Videos, CVPR 2018. [3] L. Chen, et al., MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features, CVPR 2018. [2] [3]
  • 10. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. GANNoise Text-to-image [3] (and Image-to-text)[1] [2] [1] P. Isola, et al., Image-to-Image Translation with Conditional Adversarial Nets, CVPR 2017. [2] C. Ledig, et al., Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network, CVPR 2017. [3] S. Reed, et al., Generative Adversarial Text to Image Synthesis, ICML 2016.
  • 11. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Saliency ( ) [1] [1] N. Liu, et al., PiCANet: Learning Pixel-wise Contextual Attention for Saliency Detection, CVPR 2018. [2] Z. Li, et al., MegaDepth: Learning Single-View Depth Prediction from Internet Photos, CVPR 2018. [2]
  • 12. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 0 2 4 6 8 10 12 14 16 18 20 1 2 3 4 5 6 7 8 9 1011121314151617181920 ID [1] O. Vinyals, et al., Matching Networks for One Shot Learning, arXiv:1606.04080 • • [1]
  • 13. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • Deep Learning • X. Yuan, et al., Adversarial Examples: Attacks and Defenses for Deep Learning, IEEE Trans Neural Netw Learn Syst. 2019.
  • 14. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 15. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • • • ONNX AutoML Define-by-run
  • 16. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • TensorFlow models TF slim GluonCV ChainerCV PyTorchCV
  • 17. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ResNet (Gluon vs MXNet) num_unit = len(units) assert(num_unit == num_stages) data = mx.sym.Variable(name='data') if dtype == 'float32': data = mx.sym.identity(data=data, name='id') else: if dtype == 'float16': data = mx.sym.Cast(data=data, dtype=np.float16) data = mx.sym.BatchNorm(data=data, fix_gamma=True, eps=2e-5, momentum=bn_mom, name='bn_data') (nchannel, height, width) = image_shape if height <= 32: # such as cifar10 body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(3, 3), stride=(1,1), pad=(1, 1), no_bias=True, name="conv0", workspace=workspace) else: # often expected to be 224 such as imagenet body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(7, 7), stride=(2,2), pad=(3, 3), no_bias=True, name="conv0", workspace=workspace) body = mx.sym.BatchNorm(data=body, fix_gamma=False, eps=2e-5, momentum=bn_mom, name='bn0') body = mx.sym.Activation(data=body, act_type='relu', name='relu0') body = mx.sym.Pooling(data=body, kernel=(3, 3), stride=(2,2), pad=(1,1), pool_type='max') for i in range(num_stages): body = residual_unit(body, filter_list[i+1], (1 if i==0 else 2, 1 if i==0 else 2), False, name='stage%d_unit%d' % (i + 1, 1), bottle_neck=bottle_neck, workspace=workspace, memonger=memonger) for j in range(units[i]-1): body = residual_unit(body, filter_list[i+1], (1,1), True, name='stage%d_unit%d' % (i + 1, j + 2), bottle_neck=bottle_neck, workspace=workspace, memonger=memonger) bn1 = mx.sym.BatchNorm(data=body, fix_gamma=False, eps=2e-5, momentum=bn_mom, name='bn1') relu1 = mx.sym.Activation(data=bn1, act_type='relu', name='relu1') MXNet from mxnet.gluon.model_zoo import vision resnet18 = vision.resnet18_v1() Gluon
  • 18. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ONNX (Open Neural Network Exchange) MXNet Caffe2 PyTorch TF CNTKCoreML Tensor RT NGraph SNPE • ONNX ONNX •
  • 19. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ONNX Protocol Buffers • • • API Protocol Buffers Graph Operator Tensor, … Operator Definitions ONNX Python API
  • 20. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Define-and-run Define-by-run • Define-and-run • • TensorFlow, MXNet • Define-by-run • • Chainer PyTorch, TensorFlow, MXNet
  • 21. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Define-and-run Define-by-run Define-and-run Define-by-run def our_function(A, B): C = A + B return C A = Load_Data_A() B = Load_Data_B() result = our_function(A, B) A = placeholder() B = placeholder() C = A + B our_function = compile(inputs=[A, B], outputs =[C]) A = Load_Data_A() B = Load_Data_B() result = our_function(A, B) https://gluon.mxnet.io/chapter07_distributed-learning/hybridize.html Define Run Define, Run
  • 22. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Define-by-run Define-and-run Define-by-run def our_function(A, B): C = A + B return C A = Load_Data_A() B = Load_Data_B() result = our_function(A, B) A = placeholder() B = placeholder() C = A + B our_function = compile(inputs=[A, B], outputs =[C]) A = Load_Data_A() B = Load_Data_B() result = our_function(A, B)
  • 23. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML • • , etc. D. Bayor, et al., TFX: A TensorFlow-Based Production-Scale Machine Learning Platform, KDD 2017.
  • 24. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML • AutoML • ICML 2014 AutoML * • • • Meta-Learning, Learning to learn * https://sites.google.com/site/automlwsicml14/
  • 25. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML
  • 26. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML Amazon Forecast User CSV file 1. S3 2. Forecast 3. Forecast 4.
  • 27. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 28. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • Model Server Interpretable ML
  • 29. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Model Server • • Model Server • • REST/RPC Model Server Mobile client Deploy REST/RPC
  • 30. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. TensorFlow Serving [1] C. Olston, et al., TensorFlow-Serving: Flexible, High-Performance ML Serving, NIPS 2017. • Controller, Synchronizer Serving job • Router Serving job
  • 31. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. MXNet Model Server https://aws.amazon.com/jp/blogs/news/model-server-for-apache-mxnet-v1-0-released/ • REST API • MMS 1.0 1,000 MMS 1.0 MMS 0.4
  • 32. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • • • AWS, SageMaker Neo • Nvidia, TensorRT Raspberry Pi ResNet18 Mobilenet 11.5x 2.2x
  • 33. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. SageMaker Neo / TVM • Operator Fusion • Data Layout Transformation 4x4 4x4 • Tensor Expression and Schedule Space • Nested Parallelism with Cooperation • etc… T. Chen, et al., TVM: An Automated End-to-End Optimizing Compiler for Deep Learning, OSDI 2018.
  • 34. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. TensorRT • Layer & Tensor Fusion 1 • FP16 and INT8 Precision Calibration FP32 FP16 INT8 • Kernel Auto-Tuning • Dynamic Tensor Memory • Multi Stream Execution https://devblogs.nvidia.com/tensorrt-3-faster-tensorflow-inference/
  • 35. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Interpretable ML: : SVM GBT C. Molnar, Interpretable Machine Learning, https://christophm.github.io/interpretable-ml-book/ >900< 900 < 2000 km2 > 2000 km2
  • 36. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Interpretable ML for computer vision • • M.T. Ribeiro, et al., Anchors: High-Precision Model-Agnostic Explanations, AAAI 2018.
  • 37. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 38. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • • • 1 1 • • AWS Inferentia • Intel Nervana
  • 39. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Machine Learning on FPGA • FPGA • AWS F1 instance Amazon Machine Image • Loop tiling [1] [1] C. Zhang, et al., Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks, FPGA 2015.
  • 40. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • GPU
  • 41. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 42. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 43. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • AutoML AI •
  • 44. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://amzn.to/aws_dev