SlideShare una empresa de Scribd logo

“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Presentation from AMD

•
•
Edge AI and Vision Alliance
Edge AI and Vision Alliance

For the full video of this presentation, please visit: https://www.edge-ai-vision.com/2021/09/an-introduction-to-data-augmentation-techniques-in-ml-frameworks-a-presentation-from-amd/ Rajy Rawther, PMTS Software Architect at AMD, presents the “Introduction to Data Augmentation Techniques in ML Frameworks” tutorial at the May 2021 Embedded Vision Summit. Data augmentation is a set of techniques that expand the diversity of data available for training machine learning models by generating new data from existing data. This talk introduces different types of data augmentation techniques as well as their uses in various training scenarios. Rawther explores some built-in augmentation methods in popular ML frameworks like PyTorch and TensorFlow. She also discusses some tips and tricks that are commonly used to randomly select parameters to avoid having model overfit to a particular dataset.

Tecnología
1 de 32
Descargar para leer sin conexión
© 2021 Advanced Micro Devices An Analysis of Data Augmentation Techniques in Machine Learning Frameworks Rajy Rawther Advanced Micro Devices, Inc. May 2021
© 2021 Advanced Micro Devices Agenda • Why do we need augmentation? • Different types of augmentations • Analysis of augmentations for classification vs object detection • Random parameter adjustment to prevent overfitting: Common techniques • How ML learning frameworks handle data augmentation in the training pipeline • Challenges • Closing remarks 2
© 2021 Advanced Micro Devices • In 2015, many of us have failed to correctly identify the color of this viral dress. • However, today’s neural networks can correctly predict the color of the dress with ~90% accuracy. • This is achieved by training an image classification network with a large amount of data. • Neural networks don’t have misperceptions of data, but it can learn from poor data. • The amount of data needed to train is linearly proportional to the complexity of your model. How do I get more data if I don’t have “more data”? Why Do We Need Augmentation? 3 .
© 2021 Advanced Micro Devices Overfitting 4 • ImageNet(ILSVRC 2012-2017) has 1.2 million training images, 50,000 validation images and 150,000 test images • Typical image classification convolution network has millions of parameters and thousands of neurons to train Overfitting! Overfitting occurs when the model fits too much to the training data to the extent that it performs poorly on unseen data.
© 2021 Advanced Micro Devices Example Of Overfitting 5 0 5 10 15 20 25 30 35 40 0 2 4 6 8 10 12 Classification error EPOCHS Signs of overfitting Testing Error Training Error 0 5 10 15 20 25 30 35 40 0 2 4 6 8 10 12 Classification Error EPOCHS Desired convergence Testing Error Training Error Best fit
© 2021 Advanced Micro Devices Data Augmentation Advantages • Reduce overfitting – • You don’t want the network to memorize the features of training data. • Faster training • Increased accuracy • Increased dataset size • Makes your trained neural network invariant to different aspects • Translation • Size • Illumination • location • Mask • Add hard-to-get or rare variations to the dataset 6
© 2021 Advanced Micro Devices History: AlexNet Augmentation for Classification • Dataset size is increased to 2048x by • Randomly cropping 224x224 patches • Doing color augmentations • Randomly flipping the images horizontally • Randomly resizing them 7 Resulted in 1% error rate reduction
© 2021 Advanced Micro Devices Understanding Different Use Cases For Augmentation 8 Image Classification Object Detection Segmentation Cat or Dog? Entire Image Classify Objects with Location Classify each pixel to a class
© 2021 Advanced Micro Devices Data Augmentation Categories 9 Color Geometric Filtering Mixing Effects Brightness Contrast Saturation Blur Hue Color-Temp Vignette Rain Snow Glitch Noise Fog Gamma- Correction Water Flip Resize Crop Warp Affine Scale Rotate Warp Perspective Box Sobel Gaussian Noise Median Blend Non_linear- Blend Crop And Patch Erase RICAP
© 2021 Advanced Micro Devices Color & Illumination Examples 10 Saturation Color Temp- Vignette Original Brightness Contrast
© 2021 Advanced Micro Devices Geometric and Displacement Distortion Examples Original Rotate Vertical Flip Warp Fish-Eye Effect Horizontal Flip Crop Resize 11
© 2021 Advanced Micro Devices Original Nonlinear Blend Crop And Patch Blend Glitch Water ColorTwist Erase 12 Mixing Augmentations: Disruptive
© 2021 Advanced Micro Devices Not All Augmentations Apply To All Datasets Original Rotate 90 Rotate 180 Flip (mirror) 13
© 2021 Advanced Micro Devices • Randomness in color augmentation • Real world data can exist in a variety of conditions, like low lighting, grasslands, rain, snow, etc. • Random parameter adjustments can help to overcome this by generating new data on the fly. Random Parameter Adjustments To Prevent Overfitting α = 1.0 + random.uniform(-strength, strength) Image *= α Strength to control brightness 14 Brightness Variation Noise variation
© 2021 Advanced Micro Devices By doing random geometric augmentations like scale, resize, rotate, flip, etc., you are training your network to be invariant to geometric distortions. Random Geometric Distortion Tesla Ford Ford Dataset Tesla Car – Label 0 Ford Car – Label 1 Label - 0 Tesla (wrong) Trained Neural Net 15
© 2021 Advanced Micro Devices Bounding Box Augmentations 16 Resize Flip Rotate Crop Color Augmentations don’t impact the locations of bounding boxes whereas geometric operations alters bounding box locations.
© 2021 Advanced Micro Devices Segmentation Mask Augmentations • Examples of augmentations applied to base image and mask image. 17 Image Mask Original Flip Color Twist Crop
© 2021 Advanced Micro Devices How To Build A Training Pipeline With Augmentation? Load & Decode CPU/GPU decode Training dataset Augmentations CPU/GPU based Ready to train data Training or Inference 18
© 2021 Advanced Micro Devices Image Classification Training With Augmentation Pipeline Data Iterator Output Tensor 19 Image Augmentations Data Reader Decode Metadata Reader Metadata Augmentations Dataset Labels Dataset Class Crop, Mirror Normalize … Resize 2, 100, 10, .. Mask (x,y,w,h) BBox Mask Dataset Class
© 2021 Advanced Micro Devices Augmentations In PyTorch PyTorch uses torchvision.transforms library to apply image transformations 20
© 2021 Advanced Micro Devices Augmentation Functions In Pytorch And TensorFlow 21 PyTorch (torchvision.transforms) PyTorch (torchvision.functional) TensorFlow (lambda functions) CenterCrop adjust_brightness Center_crop Normalize adjust_hue Random_brightness Resize, Scale crop Random_contrast RandomCrop equalize Random_hue ColorJitter hflip Random_flip_left_right RandomAffine vflip Random_flip_up_down RandomRotate, RandomFlip pad Resize_and_rescale
© 2021 Advanced Micro Devices A batch of images are processed in a pipeline using CPU/GPU to generate output tensor Data Augmentation Pipeline: Offline vs On The Fly A batch of images are generated from a single image using a pipeline GPU Aug 1 3 2 4 6 8 7 5 Output tensor CPU Aug Offline On the fly Augmentations
© 2021 Advanced Micro Devices Example: SSD Object Detection Training • Object detection and classification are done in a single forward pass of the network • Bounding boxes need to be processed along with images to compute the loss function • Known to perform worse on smaller objects since they disappear in some feature maps, because priors were precomputed at different feature maps. • SSD uses VGG-16 as the base network for classification • To alleviate this, SSDRandomCrop augmentation is used. L Total = Lconfidence + 𝞪 Lloc 23
© 2021 Advanced Micro Devices Example: SSD Object Detection Training Augmentations SSDRandomCrop Resize ColorTwist RandomMirror Image with bboxes Bad crop Good crop Color Twisted Resized Randomly flipped 24
© 2021 Advanced Micro Devices Data Augmentation Results Augmentation Variation Train Accuracy (top1) Validation Accuracy (top 1) Train Loss Validation Loss Almost no augmentation 91.5 65.3 1.08 2.17 Normalization 91.5 71.78 1.109 2.04 Random Resize + Random Crop + Normalization 97.7 77.2 1.5 1.65 Random Resize + Random CMN* 97.8 76.7 1.49 1.62 25 *CMN: Random Crop and Mirror with Normalization Table generated for ResNet50 training on a smaller subset of ImageNet dataset using PyTorch
© 2021 Advanced Micro Devices Training Results
© 2021 Advanced Micro Devices Data Augmentation Framework Challenges • Each Framework has its own data loader and augmentation pipelines. • Extra effort to optimize them individually: not portable Dataset Transforms Utils Torch data class Torchvision ImageInputOp Proto, LMDB CreateDB Data loading Transformations TF Dataset TF Session Map or lambda function Transformations Data loading Need a unified library which can work across all the frameworks.
© 2021 Advanced Micro Devices Data Augmentation Algorithm Challenges • Designing an ideal augmentation strategy is heuristic and can result in sub-optimal training outcomes • Data augmentation techniques can greatly depend on the dataset: e.g., face detection dataset • Component invariant (hairstyle, makeup, accessory) • Attribute (pose, expression, age) • Each of the augmentation use cases has many challenges when it comes to choosing the ideal augmentation strategy. • A unified augmentation library that can work across all frameworks can provide both performance and flexibility
© 2021 Advanced Micro Devices Wrap it up • Data Augmentation: To prevent overfitting and expand dataset • Data Augmentation pipelines can greatly vary based on the use case • Choosing the ideal augmentation pipeline is tricky and needs some automation • Neural Style Transfer and GANs bring an artistic approach to augmentation and can provide automation
© 2021 Advanced Micro Devices References 30 A survey on Image Data Augmentation for Deep Learning https://journalofbigdata.springeropen.com/articles/10. 1186/s40537-019-0197-0 rocAL (ROCm Augmentation Library) https://github.com/GPUOpen-ProfessionalCompute- Libraries/MIVisionX/tree/master/rocAL ImageNet http://www.image-net.org/challenges/LSVRC/2012/ PyTorch Transforms https://pytorch.org/vision/stable/transforms.html TensorFlow Augmentations https://www.tensorflow.org/tutorials/images/data_augme ntation MIVisionX https://github.com/GPUOpen-ProfessionalCompute- Libraries/MIVisionX
© 2021 Advanced Micro Devices Disclaimer • The information presented in this document is for informational purposes only and may contain technical inaccuracies, omissions, and typographical errors. The information contained herein is subject to change and may be rendered inaccurate for many reasons, including but not limited to product and roadmap changes, component and motherboard version changes, new model and/or product releases, product differences between differing manufacturers, software changes, BIOS flashes, firmware upgrades, or the like. Any computer system has risks of security vulnerabilities that cannot be completely prevented or mitigated. AMD assumes no obligation to update or otherwise correct or revise this information. However, AMD reserves the right to revise this information and to make changes from time to time to the content hereof without obligation of AMD to notify any person of such revisions or changes. • THIS INFORMATION IS PROVIDED ‘AS IS.” AMD MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE CONTENTS HEREOF AND ASSUMES NO RESPONSIBILITY FOR ANY INACCURACIES, ERRORS, OR OMISSIONS THAT MAY APPEAR IN THIS INFORMATION. AMD SPECIFICALLY DISCLAIMS ANY IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR ANY PARTICULAR PURPOSE. IN NO EVENT WILL AMD BE LIABLE TO ANY PERSON FOR ANY RELIANCE, DIRECT, INDIRECT, SPECIAL, OR OTHER CONSEQUENTIAL DAMAGES ARISING FROM THE USE OF ANY INFORMATION CONTAINED HEREIN, EVEN IF AMD IS EXPRESSLY ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. • © 2021 Advanced Micro Devices, Inc. All rights reserved. • AMD, the AMD Arrow logo, Epyc, Radeon, ROCm and combinations thereof are trademarks of Advanced Micro Devices, Inc. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. 31
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Presentation from AMD

Recomendados

Data Augmentation
Data AugmentationData Augmentation
Data AugmentationMd Tajul Islam
 
"Getting More from Your Datasets: Data Augmentation, Annotation and Generativ...
"Getting More from Your Datasets: Data Augmentation, Annotation and Generativ..."Getting More from Your Datasets: Data Augmentation, Annotation and Generativ...
"Getting More from Your Datasets: Data Augmentation, Annotation and Generativ...Edge AI and Vision Alliance
 
Deep Learning for Computer Vision: Data Augmentation (UPC 2016)
Deep Learning for Computer Vision: Data Augmentation (UPC 2016)Deep Learning for Computer Vision: Data Augmentation (UPC 2016)
Deep Learning for Computer Vision: Data Augmentation (UPC 2016)Universitat Politècnica de Catalunya
 
PR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsPR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsJinwon Lee
 
Image segmentation with deep learning
Image segmentation with deep learningImage segmentation with deep learning
Image segmentation with deep learningAntonio Rueda-Toicen
 
Transfer Learning and Fine-tuning Deep Neural Networks
Transfer Learning and Fine-tuning Deep Neural Networks Transfer Learning and Fine-tuning Deep Neural Networks
Transfer Learning and Fine-tuning Deep Neural NetworksPyData
 
Deep Learning Explained
Deep Learning ExplainedDeep Learning Explained
Deep Learning ExplainedMelanie Swan
 
Simple Introduction to AutoEncoder
Simple Introduction to AutoEncoderSimple Introduction to AutoEncoder
Simple Introduction to AutoEncoderJun Lang
 

Recomendados

Data Augmentation
Data AugmentationData Augmentation
Data AugmentationMd Tajul Islam
 
"Getting More from Your Datasets: Data Augmentation, Annotation and Generativ...
"Getting More from Your Datasets: Data Augmentation, Annotation and Generativ..."Getting More from Your Datasets: Data Augmentation, Annotation and Generativ...
"Getting More from Your Datasets: Data Augmentation, Annotation and Generativ...Edge AI and Vision Alliance
 
Deep Learning for Computer Vision: Data Augmentation (UPC 2016)
Deep Learning for Computer Vision: Data Augmentation (UPC 2016)Deep Learning for Computer Vision: Data Augmentation (UPC 2016)
Deep Learning for Computer Vision: Data Augmentation (UPC 2016)Universitat Politècnica de Catalunya
 
PR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsPR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsJinwon Lee
 
Image segmentation with deep learning
Image segmentation with deep learningImage segmentation with deep learning
Image segmentation with deep learningAntonio Rueda-Toicen
 
Transfer Learning and Fine-tuning Deep Neural Networks
Transfer Learning and Fine-tuning Deep Neural Networks Transfer Learning and Fine-tuning Deep Neural Networks
Transfer Learning and Fine-tuning Deep Neural NetworksPyData
 
Deep Learning Explained
Deep Learning ExplainedDeep Learning Explained
Deep Learning ExplainedMelanie Swan
 
Simple Introduction to AutoEncoder
Simple Introduction to AutoEncoderSimple Introduction to AutoEncoder
Simple Introduction to AutoEncoderJun Lang
 
GANs Presentation.pptx
GANs Presentation.pptxGANs Presentation.pptx
GANs Presentation.pptxMAHMOUD729246
 
Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)Appsilon Data Science
 
Convolutional neural network
Convolutional neural network Convolutional neural network
Convolutional neural network Yan Xu
 
Introduction to Deep learning
Introduction to Deep learningIntroduction to Deep learning
Introduction to Deep learningleopauly
 
Introduction to Deep learning
Introduction to Deep learningIntroduction to Deep learning
Introduction to Deep learningMassimiliano Ruocco
 
Transfer Learning
Transfer LearningTransfer Learning
Transfer LearningHichem Felouat
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]Dongmin Choi
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and ApplicationsEmanuele Ghelfi
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning TutorialAmr Rashed
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer VisionSungjoon Choi
 
U-Net (1).pptx
U-Net (1).pptxU-Net (1).pptx
U-Net (1).pptxChangjin Lee
 
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AIIntro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AISeth Grimes
 
Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Suraj Aavula
 
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...Simplilearn
 
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnIntroduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnBenjamin Bengfort
 
Introduction to CNN
Introduction to CNNIntroduction to CNN
Introduction to CNNShuai Zhang
 
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...Simplilearn
 
Deep Learning - Convolutional Neural Networks
Deep Learning - Convolutional Neural NetworksDeep Learning - Convolutional Neural Networks
Deep Learning - Convolutional Neural NetworksChristian Perone
 
Federated learning
Federated learningFederated learning
Federated learningMindos Cheng
 
Transfer learning-presentation
Transfer learning-presentationTransfer learning-presentation
Transfer learning-presentationBushra Jbawi
 
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...Edge AI and Vision Alliance
 
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIAEdge AI and Vision Alliance
 

Más contenido relacionado

La actualidad más candente

GANs Presentation.pptx
GANs Presentation.pptxGANs Presentation.pptx
GANs Presentation.pptxMAHMOUD729246
 
Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)Appsilon Data Science
 
Convolutional neural network
Convolutional neural network Convolutional neural network
Convolutional neural network Yan Xu
 
Introduction to Deep learning
Introduction to Deep learningIntroduction to Deep learning
Introduction to Deep learningleopauly
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]Dongmin Choi
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and ApplicationsEmanuele Ghelfi
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning TutorialAmr Rashed
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer VisionSungjoon Choi
 
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AIIntro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AISeth Grimes
 
Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Suraj Aavula
 
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...Simplilearn
 
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnIntroduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnBenjamin Bengfort
 
Introduction to CNN
Introduction to CNNIntroduction to CNN
Introduction to CNNShuai Zhang
 
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...Simplilearn
 
Deep Learning - Convolutional Neural Networks
Deep Learning - Convolutional Neural NetworksDeep Learning - Convolutional Neural Networks
Deep Learning - Convolutional Neural NetworksChristian Perone
 
Federated learning
Federated learningFederated learning
Federated learningMindos Cheng
 
Transfer learning-presentation
Transfer learning-presentationTransfer learning-presentation
Transfer learning-presentationBushra Jbawi
 

La actualidad más candente (20)

GANs Presentation.pptx
GANs Presentation.pptxGANs Presentation.pptx
GANs Presentation.pptx
 
Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)
 
Convolutional neural network
Convolutional neural network Convolutional neural network
Convolutional neural network
 
Introduction to Deep learning
Introduction to Deep learningIntroduction to Deep learning
Introduction to Deep learning
 
Introduction to Deep learning
Introduction to Deep learningIntroduction to Deep learning
Introduction to Deep learning
 
Transfer Learning
Transfer LearningTransfer Learning
Transfer Learning
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and Applications
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning Tutorial
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer Vision
 
U-Net (1).pptx
U-Net (1).pptxU-Net (1).pptx
U-Net (1).pptx
 
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AIIntro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
Intro to Deep Learning for Medical Image Analysis, with Dan Lee from Dentuit AI
 
Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)
 
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
Recurrent Neural Network (RNN) | RNN LSTM Tutorial | Deep Learning Course | S...
 
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-LearnIntroduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-Learn
 
Introduction to CNN
Introduction to CNNIntroduction to CNN
Introduction to CNN
 
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
What Is Deep Learning? | Introduction to Deep Learning | Deep Learning Tutori...
 
Deep Learning - Convolutional Neural Networks
Deep Learning - Convolutional Neural NetworksDeep Learning - Convolutional Neural Networks
Deep Learning - Convolutional Neural Networks
 
Federated learning
Federated learningFederated learning
Federated learning
 
Transfer learning-presentation
Transfer learning-presentationTransfer learning-presentation
Transfer learning-presentation
 

Similar a “An Introduction to Data Augmentation Techniques in ML Frameworks,” a Presentation from AMD

“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...Edge AI and Vision Alliance
 
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIAEdge AI and Vision Alliance
 
CAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGCAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGIRJET Journal
 
“Automated Neural Network Model Training: The Impact on Deploying and Scaling...
“Automated Neural Network Model Training: The Impact on Deploying and Scaling...“Automated Neural Network Model Training: The Impact on Deploying and Scaling...
“Automated Neural Network Model Training: The Impact on Deploying and Scaling...Edge AI and Vision Alliance
 
“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...
“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...
“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...Edge AI and Vision Alliance
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用CHENHuiMei
 
IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...
IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...
IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...IRJET Journal
 
"Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr...
"Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr..."Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr...
"Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr...Edge AI and Vision Alliance
 
Computer Vision for Beginners
Computer Vision for BeginnersComputer Vision for Beginners
Computer Vision for BeginnersSanghamitra Deb
 
“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...
“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...
“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...Edge AI and Vision Alliance
 
“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...
“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...
“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...Edge AI and Vision Alliance
 
“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...
“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...
“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...Edge AI and Vision Alliance
 
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakLearn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakPyData
 
“Applying the Right Deep Learning Model with the Right Data for Your Applicat...
“Applying the Right Deep Learning Model with the Right Data for Your Applicat...“Applying the Right Deep Learning Model with the Right Data for Your Applicat...
“Applying the Right Deep Learning Model with the Right Data for Your Applicat...Edge AI and Vision Alliance
 
Microsoft_Databricks Datathon - Submission Deck TEMPLATE.pptx
Microsoft_Databricks Datathon - Submission Deck TEMPLATE.pptxMicrosoft_Databricks Datathon - Submission Deck TEMPLATE.pptx
Microsoft_Databricks Datathon - Submission Deck TEMPLATE.pptxAbdoulaye DOUCOURE
 
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...Jinwon Lee
 
Graph Data Science at Scale
Graph Data Science at ScaleGraph Data Science at Scale
Graph Data Science at ScaleNeo4j
 
PR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked AutoencodersPR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked AutoencodersSunghoon Joo
 
“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...
“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...
“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...Edge AI and Vision Alliance
 
Data Quality for Machine Learning Tasks
Data Quality for Machine Learning TasksData Quality for Machine Learning Tasks
Data Quality for Machine Learning TasksHima Patel
 

Similar a “An Introduction to Data Augmentation Techniques in ML Frameworks,” a Presentation from AMD (20)

“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
“Practical Image Data Augmentation Methods for Training Deep Learning Object ...
 
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
“Getting Started with Vision AI Model Training,” a Presentation from NVIDIA
 
CAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGCAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNING
 
“Automated Neural Network Model Training: The Impact on Deploying and Scaling...
“Automated Neural Network Model Training: The Impact on Deploying and Scaling...“Automated Neural Network Model Training: The Impact on Deploying and Scaling...
“Automated Neural Network Model Training: The Impact on Deploying and Scaling...
 
“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...
“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...
“DNN Training Data: How to Know What You Need and How to Get It,” a Presentat...
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用
 
IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...
IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...
IRJET - Multi-Label Road Scene Prediction for Autonomous Vehicles using Deep ...
 
"Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr...
"Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr..."Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr...
"Enabling Ubiquitous Visual Intelligence Through Deep Learning," a Keynote Pr...
 
Computer Vision for Beginners
Computer Vision for BeginnersComputer Vision for Beginners
Computer Vision for Beginners
 
“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...
“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...
“High-fidelity Conversion of Floating-point Networks for Low-precision Infere...
 
“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...
“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...
“Tools and Strategies for Quickly Building Effective Image Datasets,” a Prese...
 
“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...
“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...
“Deep Learning for Manufacturing Inspection: Case Studies,” a Presentation fr...
 
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakLearn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
 
“Applying the Right Deep Learning Model with the Right Data for Your Applicat...
“Applying the Right Deep Learning Model with the Right Data for Your Applicat...“Applying the Right Deep Learning Model with the Right Data for Your Applicat...
“Applying the Right Deep Learning Model with the Right Data for Your Applicat...
 
Microsoft_Databricks Datathon - Submission Deck TEMPLATE.pptx
Microsoft_Databricks Datathon - Submission Deck TEMPLATE.pptxMicrosoft_Databricks Datathon - Submission Deck TEMPLATE.pptx
Microsoft_Databricks Datathon - Submission Deck TEMPLATE.pptx
 
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
 
Graph Data Science at Scale
Graph Data Science at ScaleGraph Data Science at Scale
Graph Data Science at Scale
 
PR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked AutoencodersPR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked Autoencoders
 
“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...
“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...
“Fundamentals of Training AI Models for Computer Vision Applications,” a Pres...
 
Data Quality for Machine Learning Tasks
Data Quality for Machine Learning TasksData Quality for Machine Learning Tasks
Data Quality for Machine Learning Tasks
 

Más de Edge AI and Vision Alliance

“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...Edge AI and Vision Alliance
 
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...Edge AI and Vision Alliance
 
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...Edge AI and Vision Alliance
 
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...Edge AI and Vision Alliance
 
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...Edge AI and Vision Alliance
 
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...Edge AI and Vision Alliance
 
“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...Edge AI and Vision Alliance
 
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsightsEdge AI and Vision Alliance
 
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...Edge AI and Vision Alliance
 
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...Edge AI and Vision Alliance
 
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...Edge AI and Vision Alliance
 
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...Edge AI and Vision Alliance
 
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...Edge AI and Vision Alliance
 
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...Edge AI and Vision Alliance
 
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...Edge AI and Vision Alliance
 
“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from Samsara“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from SamsaraEdge AI and Vision Alliance
 
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...Edge AI and Vision Alliance
 
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...Edge AI and Vision Alliance
 
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...Edge AI and Vision Alliance
 
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...Edge AI and Vision Alliance
 

Más de Edge AI and Vision Alliance (20)

“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
 
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
 
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
 
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
 
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
 
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
 
“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...
 
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
 
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
 
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
 
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
 
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
 
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
 
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
 
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
 
“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from Samsara“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from Samsara
 
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
 
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
 
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
 
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
 

Último

DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDropbox
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native ApplicationsWSO2
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsNanddeep Nachan
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024The Digital Insurer
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfOverkill Security
 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024The Digital Insurer
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MIND CTI
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxRustici Software
 
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Zilliz
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024The Digital Insurer
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
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.pdfsudhanshuwaghmare1
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 

Último (20)

DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
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
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 

“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Presentation from AMD

  • 1. © 2021 Advanced Micro Devices An Analysis of Data Augmentation Techniques in Machine Learning Frameworks Rajy Rawther Advanced Micro Devices, Inc. May 2021
  • 2. © 2021 Advanced Micro Devices Agenda • Why do we need augmentation? • Different types of augmentations • Analysis of augmentations for classification vs object detection • Random parameter adjustment to prevent overfitting: Common techniques • How ML learning frameworks handle data augmentation in the training pipeline • Challenges • Closing remarks 2
  • 3. © 2021 Advanced Micro Devices • In 2015, many of us have failed to correctly identify the color of this viral dress. • However, today’s neural networks can correctly predict the color of the dress with ~90% accuracy. • This is achieved by training an image classification network with a large amount of data. • Neural networks don’t have misperceptions of data, but it can learn from poor data. • The amount of data needed to train is linearly proportional to the complexity of your model. How do I get more data if I don’t have “more data”? Why Do We Need Augmentation? 3 .
  • 4. © 2021 Advanced Micro Devices Overfitting 4 • ImageNet(ILSVRC 2012-2017) has 1.2 million training images, 50,000 validation images and 150,000 test images • Typical image classification convolution network has millions of parameters and thousands of neurons to train Overfitting! Overfitting occurs when the model fits too much to the training data to the extent that it performs poorly on unseen data.
  • 5. © 2021 Advanced Micro Devices Example Of Overfitting 5 0 5 10 15 20 25 30 35 40 0 2 4 6 8 10 12 Classification error EPOCHS Signs of overfitting Testing Error Training Error 0 5 10 15 20 25 30 35 40 0 2 4 6 8 10 12 Classification Error EPOCHS Desired convergence Testing Error Training Error Best fit
  • 6. © 2021 Advanced Micro Devices Data Augmentation Advantages • Reduce overfitting – • You don’t want the network to memorize the features of training data. • Faster training • Increased accuracy • Increased dataset size • Makes your trained neural network invariant to different aspects • Translation • Size • Illumination • location • Mask • Add hard-to-get or rare variations to the dataset 6
  • 7. © 2021 Advanced Micro Devices History: AlexNet Augmentation for Classification • Dataset size is increased to 2048x by • Randomly cropping 224x224 patches • Doing color augmentations • Randomly flipping the images horizontally • Randomly resizing them 7 Resulted in 1% error rate reduction
  • 8. © 2021 Advanced Micro Devices Understanding Different Use Cases For Augmentation 8 Image Classification Object Detection Segmentation Cat or Dog? Entire Image Classify Objects with Location Classify each pixel to a class
  • 9. © 2021 Advanced Micro Devices Data Augmentation Categories 9 Color Geometric Filtering Mixing Effects Brightness Contrast Saturation Blur Hue Color-Temp Vignette Rain Snow Glitch Noise Fog Gamma- Correction Water Flip Resize Crop Warp Affine Scale Rotate Warp Perspective Box Sobel Gaussian Noise Median Blend Non_linear- Blend Crop And Patch Erase RICAP
  • 10. © 2021 Advanced Micro Devices Color & Illumination Examples 10 Saturation Color Temp- Vignette Original Brightness Contrast
  • 11. © 2021 Advanced Micro Devices Geometric and Displacement Distortion Examples Original Rotate Vertical Flip Warp Fish-Eye Effect Horizontal Flip Crop Resize 11
  • 12. © 2021 Advanced Micro Devices Original Nonlinear Blend Crop And Patch Blend Glitch Water ColorTwist Erase 12 Mixing Augmentations: Disruptive
  • 13. © 2021 Advanced Micro Devices Not All Augmentations Apply To All Datasets Original Rotate 90 Rotate 180 Flip (mirror) 13
  • 14. © 2021 Advanced Micro Devices • Randomness in color augmentation • Real world data can exist in a variety of conditions, like low lighting, grasslands, rain, snow, etc. • Random parameter adjustments can help to overcome this by generating new data on the fly. Random Parameter Adjustments To Prevent Overfitting α = 1.0 + random.uniform(-strength, strength) Image *= α Strength to control brightness 14 Brightness Variation Noise variation
  • 15. © 2021 Advanced Micro Devices By doing random geometric augmentations like scale, resize, rotate, flip, etc., you are training your network to be invariant to geometric distortions. Random Geometric Distortion Tesla Ford Ford Dataset Tesla Car – Label 0 Ford Car – Label 1 Label - 0 Tesla (wrong) Trained Neural Net 15
  • 16. © 2021 Advanced Micro Devices Bounding Box Augmentations 16 Resize Flip Rotate Crop Color Augmentations don’t impact the locations of bounding boxes whereas geometric operations alters bounding box locations.
  • 17. © 2021 Advanced Micro Devices Segmentation Mask Augmentations • Examples of augmentations applied to base image and mask image. 17 Image Mask Original Flip Color Twist Crop
  • 18. © 2021 Advanced Micro Devices How To Build A Training Pipeline With Augmentation? Load & Decode CPU/GPU decode Training dataset Augmentations CPU/GPU based Ready to train data Training or Inference 18
  • 19. © 2021 Advanced Micro Devices Image Classification Training With Augmentation Pipeline Data Iterator Output Tensor 19 Image Augmentations Data Reader Decode Metadata Reader Metadata Augmentations Dataset Labels Dataset Class Crop, Mirror Normalize … Resize 2, 100, 10, .. Mask (x,y,w,h) BBox Mask Dataset Class
  • 20. © 2021 Advanced Micro Devices Augmentations In PyTorch PyTorch uses torchvision.transforms library to apply image transformations 20
  • 21. © 2021 Advanced Micro Devices Augmentation Functions In Pytorch And TensorFlow 21 PyTorch (torchvision.transforms) PyTorch (torchvision.functional) TensorFlow (lambda functions) CenterCrop adjust_brightness Center_crop Normalize adjust_hue Random_brightness Resize, Scale crop Random_contrast RandomCrop equalize Random_hue ColorJitter hflip Random_flip_left_right RandomAffine vflip Random_flip_up_down RandomRotate, RandomFlip pad Resize_and_rescale
  • 22. © 2021 Advanced Micro Devices A batch of images are processed in a pipeline using CPU/GPU to generate output tensor Data Augmentation Pipeline: Offline vs On The Fly A batch of images are generated from a single image using a pipeline GPU Aug 1 3 2 4 6 8 7 5 Output tensor CPU Aug Offline On the fly Augmentations
  • 23. © 2021 Advanced Micro Devices Example: SSD Object Detection Training • Object detection and classification are done in a single forward pass of the network • Bounding boxes need to be processed along with images to compute the loss function • Known to perform worse on smaller objects since they disappear in some feature maps, because priors were precomputed at different feature maps. • SSD uses VGG-16 as the base network for classification • To alleviate this, SSDRandomCrop augmentation is used. L Total = Lconfidence + 𝞪 Lloc 23
  • 24. © 2021 Advanced Micro Devices Example: SSD Object Detection Training Augmentations SSDRandomCrop Resize ColorTwist RandomMirror Image with bboxes Bad crop Good crop Color Twisted Resized Randomly flipped 24
  • 25. © 2021 Advanced Micro Devices Data Augmentation Results Augmentation Variation Train Accuracy (top1) Validation Accuracy (top 1) Train Loss Validation Loss Almost no augmentation 91.5 65.3 1.08 2.17 Normalization 91.5 71.78 1.109 2.04 Random Resize + Random Crop + Normalization 97.7 77.2 1.5 1.65 Random Resize + Random CMN* 97.8 76.7 1.49 1.62 25 *CMN: Random Crop and Mirror with Normalization Table generated for ResNet50 training on a smaller subset of ImageNet dataset using PyTorch
  • 26. © 2021 Advanced Micro Devices Training Results
  • 27. © 2021 Advanced Micro Devices Data Augmentation Framework Challenges • Each Framework has its own data loader and augmentation pipelines. • Extra effort to optimize them individually: not portable Dataset Transforms Utils Torch data class Torchvision ImageInputOp Proto, LMDB CreateDB Data loading Transformations TF Dataset TF Session Map or lambda function Transformations Data loading Need a unified library which can work across all the frameworks.
  • 28. © 2021 Advanced Micro Devices Data Augmentation Algorithm Challenges • Designing an ideal augmentation strategy is heuristic and can result in sub-optimal training outcomes • Data augmentation techniques can greatly depend on the dataset: e.g., face detection dataset • Component invariant (hairstyle, makeup, accessory) • Attribute (pose, expression, age) • Each of the augmentation use cases has many challenges when it comes to choosing the ideal augmentation strategy. • A unified augmentation library that can work across all frameworks can provide both performance and flexibility
  • 29. © 2021 Advanced Micro Devices Wrap it up • Data Augmentation: To prevent overfitting and expand dataset • Data Augmentation pipelines can greatly vary based on the use case • Choosing the ideal augmentation pipeline is tricky and needs some automation • Neural Style Transfer and GANs bring an artistic approach to augmentation and can provide automation
  • 30. © 2021 Advanced Micro Devices References 30 A survey on Image Data Augmentation for Deep Learning https://journalofbigdata.springeropen.com/articles/10. 1186/s40537-019-0197-0 rocAL (ROCm Augmentation Library) https://github.com/GPUOpen-ProfessionalCompute- Libraries/MIVisionX/tree/master/rocAL ImageNet http://www.image-net.org/challenges/LSVRC/2012/ PyTorch Transforms https://pytorch.org/vision/stable/transforms.html TensorFlow Augmentations https://www.tensorflow.org/tutorials/images/data_augme ntation MIVisionX https://github.com/GPUOpen-ProfessionalCompute- Libraries/MIVisionX
  • 31. © 2021 Advanced Micro Devices Disclaimer • The information presented in this document is for informational purposes only and may contain technical inaccuracies, omissions, and typographical errors. The information contained herein is subject to change and may be rendered inaccurate for many reasons, including but not limited to product and roadmap changes, component and motherboard version changes, new model and/or product releases, product differences between differing manufacturers, software changes, BIOS flashes, firmware upgrades, or the like. Any computer system has risks of security vulnerabilities that cannot be completely prevented or mitigated. AMD assumes no obligation to update or otherwise correct or revise this information. However, AMD reserves the right to revise this information and to make changes from time to time to the content hereof without obligation of AMD to notify any person of such revisions or changes. • THIS INFORMATION IS PROVIDED ‘AS IS.” AMD MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE CONTENTS HEREOF AND ASSUMES NO RESPONSIBILITY FOR ANY INACCURACIES, ERRORS, OR OMISSIONS THAT MAY APPEAR IN THIS INFORMATION. AMD SPECIFICALLY DISCLAIMS ANY IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR ANY PARTICULAR PURPOSE. IN NO EVENT WILL AMD BE LIABLE TO ANY PERSON FOR ANY RELIANCE, DIRECT, INDIRECT, SPECIAL, OR OTHER CONSEQUENTIAL DAMAGES ARISING FROM THE USE OF ANY INFORMATION CONTAINED HEREIN, EVEN IF AMD IS EXPRESSLY ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. • © 2021 Advanced Micro Devices, Inc. All rights reserved. • AMD, the AMD Arrow logo, Epyc, Radeon, ROCm and combinations thereof are trademarks of Advanced Micro Devices, Inc. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. 31