AWS reinvent 2019 recap - Riyadh - AI And ML - Ahmed Raafat
0 likes540 views
Amazon Web Services provides several machine learning services and tools to help developers build, train, deploy, and manage machine learning models at scale. These include Amazon SageMaker for developing and hosting models, Amazon Rekognition for computer vision tasks, and Amazon Transcribe for speech-to-text. AWS aims to put machine learning in the hands of every developer through these fully managed services.
AWS reinvent 2019 recap - Riyadh - AI And ML - Ahmed Raafat
- 1. © 2020, Amazon Web Services, Inc. or its affiliates. All rights reserved. Ahmed Raafat-Solutions Architect arraafat@amazon.ae AWS | Middle East
- 2. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. https://aws.amazon.com/new/reinvent/
- 3. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
- 4. Why ML?!
- 5. The reach of ML is growing
- 6. 6
- 7. Our mission at AWS Put machine learning in the hands of every developer
- 8. VISION SPEECH TEXT SEARCH NEW CHATBOTS PERSONALIZATION FORECASTING FRAUD NEW DEVELOPMENT NEW CONTACT CENTERS NEW Amazon SageMaker Ground Truth Augmented AI SageMaker Neo Built-in algorithms SageMaker Notebooks NEW SageMaker Experiments NEW Model tuning SageMaker Debugger NEW SageMaker Autopilot NEW Model hosting SageMaker Model Monitor NEW Deep Learning AMIs & Containers GPUs & CPUs Elastic Inference Inferentia (Inf1) FPGA Amazon Rekognition Amazon Polly Amazon Transcribe +Medical Amazon Comprehend +Medical Amazon Translate Amazon Lex Amazon Personalize Amazon Forecast Amazon Fraud Detector Amazon CodeGuru AI SERVICES ML SERVICES ML FRAMEWORKS & INFRASTRUCTURE Amazon Textract Amazon Kendra Contact Lens For Amazon Connect SageMaker Studio IDE NEW NEW AWS Machine Learning stack NEW
- 9. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
- 10. The machine learning workflow is iterative and complex Prepare Build Train & Tune Deploy & Manage 101011010 010101010 000011110 Collect and prepare training data Choose or build an ML algorithm Set up and manage environments for training Train, debug, and tune models Deploy model in production Manage training runs Monitor models Validate predictions Scale and manage the production environment Challenging ?!
- 11. Amazon SageMaker a fully-managed platform that enables developers and data scientists to quickly and easily build, train, and deploy machine learning
- 12. Amazon SageMaker Studio The first fully integrated development environment (IDE) for machine learning Organize, track, and compare thousands of experiments Easy experiment management Share scalable notebooks without tracking code dependencies Collaboration at scale Get accurate models for with full visibility & control without writing code Automatic model generation Automatically debug errors, monitor models, & maintain high quality Higher quality ML models Code, build, train, deploy, & monitor in a unified visual interface Increased productivity
- 14. Amazon SageMaker Notebooks Access your notebooks in seconds with your corporate credentials Fast-start shareable notebooks Administrators manage access and permissions Share your notebooks as a URL with a single click Dial up or down compute resources Start your notebooks without spinning up compute resources
- 15. • You can share any of your notebooks with others in your organization. • It is a copy, so any changes you make to your notebook aren’t reflected in a previous version that you shared. • If you want to share your latest version, you must create a new snapshot and then share it. Fast-start shareable notebooks
- 16. Amazon SageMaker Processing Analytics jobs for data processing and model evaluation Use SageMaker’s built-in containers or bring your own Bring your own script for feature engineering Custom processing Achieve distributed processing for clusters Your resources are created, configured, & terminated automatically Leverage SageMaker’s security & compliance features New Python SDK that lets data scientists and ML engineers easily run preprocessing, postprocessing and model evaluation workloads on Amazon SageMaker. This SDK uses SageMaker’s built-in container for scikit-learn, possibly the most popular library one for data set transformation.
- 17. Amazon SageMaker Processing • Converting the data set to the input format expected by the ML algorithm you’re using. • Transforming existing features to a more expressive representation, such as one- hot encoding categorical features. • Rescaling or normalizing numerical features. • Engineering high level features, e.g. replacing mailing addresses with GPS coordinates. • Cleaning and tokenizing text for natural language processing applications. • And more! https://aws.amazon.com/blogs/aws/amazon-sagemaker-processing-fully-managed-data- processing-and-model-evaluation/ A processing job downloads input from Amazon Simple Storage Service (Amazon S3), then uploads outputs to Amazon S3 during or after the processing job.
- 18. Demo!
- 19. Amazon SageMaker Experiments Experiment tracking at scale Visualization for best results Flexibility with Python SDK & APIs Iterate quickly Track parameters & metrics across experiments & users Organize experiments Organize by teams, goals, & hypotheses Visualize & compare between experiments Log custom metrics & track models using APIs Iterate & develop high- quality models A system to organize, track, and evaluate training experiments
- 21. Automatic data analysis Relevant data capture Automatic error detection Improved productivity with alerts Visual analysis and debug Amazon SageMaker Debugger Analyze and debug data with no code changes Data is automatically captured for analysis Errors are automatically detected based on rules Take corrective action based on alerts Visually analyze & debug from SageMaker Studio Analysis & debugging, explainability, and alert generation The debugging effort for machine learning models from days to minutes by providing complete insights into the debugging process.
- 22. Use Amazon SageMaker Debugger to identify issues such as vanishing gradients SHAP (SHapley Additiv e exPlanations)
- 23. Amazon SageMaker Model Monitor Automatic data collection Continuous Monitoring CloudWatch Integration Data is automatically collected from your endpoints Automate corrective actions based on Amazon CloudWatch alerts Continuous monitoring of models in production Visual Data analysis Define a monitoring schedule and detect changes in quality against a pre-defined baseline See monitoring results, data statistics, and violation reports in SageMaker Studio Flexibility with rules Use built-in rules to detect data drift or write your own rules for custom analysis
- 25. Amazon SageMaker Autopilot Quick to start Provide your data in a tabular form & specify target prediction Automatic model creation Get ML models with feature engineering & automatic model tuning automatically done Visibility & control Get notebooks for your models with source code Automatic model creation with full visibility & control Recommendations & Optimization Get a leaderboard & continue to improve your model
- 27. Fully managed infrastructure in SageMaker Amazon SageMaker Operators for Kubernetes to train, tune, & deploy models
- 28. Demo!
- 29. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
- 30. VISION SPEECH TEXT SEARCH NEW CHATBOTS PERSONALIZATION FORECASTING FRAUD NEW DEVELOPMENT NEW CONTACT CENTERS NEW Amazon SageMaker Ground Truth Augmented AI SageMaker Neo Built-in algorithms SageMaker Notebooks NEW SageMaker Experiments NEW Model tuning SageMaker Debugger NEW SageMaker Autopilot NEW Model hosting SageMaker Model Monitor NEW Deep Learning AMIs & Containers GPUs & CPUs Elastic Inference Inferentia (Inf1) FPGA Amazon Rekognition Amazon Polly Amazon Transcribe +Medical Amazon Comprehend +Medical Amazon Translate Amazon Lex Amazon Personalize Amazon Forecast Amazon Fraud Detector Amazon CodeGuru AI SERVICES ML SERVICES ML FRAMEWORKS & INFRASTRUCTURE Amazon Textract Amazon Kendra Contact Lens For Amazon Connect SageMaker Studio IDE NEW NEW AWS Machine Learning stack NEW
- 31. Pre:Invent highlights https://aws.amazon.com/about-aws/whats-new/machine-learning • Amazon Comprehend: 6 new languages • Amazon Translate: 22 new languages • Amazon Transcribe: 15 new languages, alternative transcriptions • Amazon Lex: SOC and HIPAA compliance, sentiment analysis, web & mobile integration with Amazon Connect • Amazon Personalize: batch recommendations • Amazon Forecast: use any quantile for your predictions With region expansion across the board!
- 32. Amazon Fraud Detector A fraud detection service that makes it easy for businesses to use machine learning to detect online fraud in real-time, at scale
- 33. Amazon Fraud Detector – Automated Model Building 1 2 4 5 Training data in S3 63
- 34. Detect common types of online fraud Designed to help companies detect common types of online fraud Examples: • New account fraud • Online payment fraud (coming soon) • Guest checkout fraud • ‘Try Before You Buy’ + post-paid online service abuse
- 35. Generating Fraud Predictions Guest Checkout: Purchase IP: 1.23.123.123 email: joe@example.com Payment: Bank123 … Fraud Detector returns: Outcome: Approved ML Score: 160 Purchase Approved Call service with: IP: 1.23.123.123 email: joe@example.com Payment: Bank123 … AWS Cloud
- 36. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Trademark Amazon Transcribe Amazon Transcribe is a fully managed and continuously trained automatic speech recognition service powered by deep learning models. Developers can use Amazon Transcribe to easily add speech-to-text capabilities to their applications.
- 37. Amazon Transcribe Medical Easy-to-UseAccurate Affordable
- 38. Amazon Rekognition Custom Labels • Identify the objects and scenes in images that are specific to your business needs. • For example, • Find your logo in social media posts. • Identify your products on store shelves. • classify machine parts in an assembly line. • Distinguish healthy and infected plants. • Detect animated characters in videos.
- 39. Amazon Rekognition Custom Labels • Import images labeled by Amazon SageMaker Ground Truth… • Or label images automatically based on folder structure • Train a model on fully managed infrastructure • Split the data set for training and validation • See precision, recall, and F1 score at the end of training • Select your model • Use it with the usual Rekognition APIs
- 40. Demo!
- 41. 41© 2019 Amazon Web Services, Inc. or its affiliates. All rights reserved | à Still a challenge today Enterprise Search
- 42. 42© 2019 Amazon Web Services, Inc. or its affiliates. All rights reserved | Key Challenges Low Accuracy • 80% of data is unstructured • Keyword Engines Complexity • Scattered Data Silos • Stale Search Results • Difficult to set up
- 43. 43© 2019 Amazon Web Services, Inc. or its affiliates. All rights reserved | Impact on Enterprise • Lower employee productivity • Increased risk and liability • Duplication of work • Creates negative customer experience
- 44. Amazon Kendra-Rethinking Enterprise Search Find exactly what you are looking for Fast search, and quick to set up Native connectors (S3, Sharepoint, file servers, HTTP, etc.) Natural language Queries NLU and ML core Simple API and console experiences Code samples Incremental learning through feedback Domain Expertise
- 45. Ask intuitive questions Natural language queries Keyword queries
- 46. Kendra connectors …and more coming in 2020
- 48. Getting started Step 1 Create an index An index is the place where you add your data sources to make them searchable in Kendra. Step 2 Add data sources Add and sync your data from S3, Sharepoint, Box and other data sources, to your index. Step 3 Test & deploy After syncing your data, visit the Search console page to test search & deploy Kendra in your search application.
- 49. 1
- 50. 2
- 51. 3
- 52. Demo!
- 53. A2I lets you easily implement human review in machine learning workflows to improve the accuracy, speed, and scale of complex decisions. Amazon Augmented AI (A2I)
- 54. How Amazon Augmented AI works Client application sends input data AWS AI Service or custom ML model makes predictions Results stored to your S3 1 2 4 Low confidence predictions sent for human review 3 High-confidence predictions returned immediately to client application 5 Amazon Rekognition Amazon Textract
- 56. Contact Lens For Amazon Connect Theme detection Built-in automatic call transcription Automated contact categorization Enhanced Contact Search Real-time sentiment dashboard and alerting Presents recurring issues based on Customer feedback Identify call types such as script compliance, competitive mentions, and cancellations. Filter calls of interest based on words spoken and customer sentiment View entire call transcript directly in Amazon Connect Quickly identify when customers are having a poor experience on live calls Easily use the power of machine learning to improve the quality of your customer experience without requiring any technical expertise
- 58. AWS CodeGuru Built-in code reviews with intelligent recommendations Detect and optimize expensive lines of code Identify latency and performance improvements CodeGuru Reviewer CodeGuru Profiler Write + Review Build + Test Deploy Measure Improve
- 59. CodeGuru Reviewer: How It Works Input: Source Code Feature Extraction Machine Learning Output: Recommendations Customer provides source code as input Java AWS CodeCommit Github Extract semantic features / patterns ML algorithms identify similar code for comparison Customers see recommendations as Pull Request feedback
- 60. CodeGuru Example – Looping vs Waiting do { DescribeTableResult describe = ddbClient.describeTable(new DescribeTableRequest().withTableName(tableName)); String status = describe.getTable().getTableStatus(); if (TableStatus.ACTIVE.toString().equals(status)) { return describe.getTable(); } if (TableStatus.DELETING.toString().equals(status)) { throw new ResourceInUseException("Table is " + status + ", and waiting for it to become ACTIVE is not useful."); } Thread.sleep(10 * 1000); elapsedMs = System.currentTimeMillis() - startTimeMs; } while (elapsedMs / 1000.0 < waitTimeSeconds); throw new ResourceInUseException("Table did not become ACTIVE after "); This code appears to be waiting for a resource before it runs. You could use the waiters feature to help improve efficiency. Consider using TableExists, TableNotExists. For more information, see https://aws.amazon.com/blogs/developer/waiters-in-the-aws-sdk-for-java/ Recommendation Code We should use waiters instead - will help remove a lot of this code.Developer Feedback
- 61. LOWER COSTINCREASE IN CPU UTILIZATION AMAZON PRIME DAY 2017 VS 2018
- 62. Demo!
- 63. AWS DeepRacer improvements • AWS DeepRacer Evo • Stereo camera • LIDAR sensor • New racing opportunities • Create your own races • Object Detection & Avoidance • Head-to-head racing
- 64. AWS DeepComposer • MIDI keyboard to experiment with music generation using ML • Compose music using Generative Adversarial Networks (GAN) • Use a pretrained model, or train your own
- 65. Deep Graph Library https://www.dgl.ai • Python open source library that helps researchers and scientists quickly build, train, and evaluate Graph Neural Networks on their data sets • Use cases: recommendation, social networks, life sciences, cybersecurity, etc. • Available in Deep Learning Containers • PyTorch and Apache MXNet, TensorFlow coming soon • Available for training on Amazon SageMaker
- 66. Deep Java Library https://www.djl.ai • Java open source library, to train and deploy models • Framework agnostic • Apache MXNet for now, more will come • Train your own model, or use a pretrained one from the model zoo
- 67. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. https://aws.amazon.com/new/reinvent Ahmed Raafat- Solutions Architect arraafat@amazon.ae AWS | Middle East