Taking Models to the Next Level with Azure Machine Learning

Explore the integration of text analysis intelligent services into business processes using Azure Machine Learning in this conference talk from NDC Sydney 2020. Learn about pre-built cognitive services models and progress to training custom neural models for Aspect-Based Sentiment Analysis using Intel NLP Architect. Discover when custom models are necessary and how to create them quickly with AutoML. Gain insights into fine-tuning model hyperparameters using HyperDrive. Follow the journey of Tailwind Traders as they implement these technologies, addressing challenges in distributed training and managing machine learning processes. Dive into the typical end-to-end ML process, including preparation, computation, experimentation, model registration, and deployment. Understand automated hyperparameter tuning and active job management. Conclude with valuable learning resources to further your Azure ML Services and Python SDK knowledge.

Intro
The Story
Why Azure Cognitive Services
programming
Building a Baseline with AutoML
NLP-Architect Aspect Based Sentiment Analysis
Managing Tailwind Traders ML Challenges
Challenges of distributed training
Azure Machine Learning Typical end to end ML process Prepare
Create Compute
Create an experiment
Create a training file
Create an estimator
Submit the experiment to the cluster
Register the model
AMLS to deploy
Inference config inference_config - InferenceConfig
Deployment using AML
Deploy to ACI
Typical 'manual' approach to hyperparameter tu
Automated Hyperparameter Tuning Manage Active Jobs
Wrap up
Learning Resources Get started with Azure ML Services and the Python SDK aka.ms/AA3dzht