Azure Platform Solutions (Machine Learning and Cognitive Services)

Machine Learning:

Machine learning is about using existing data to forecast future behaviors outcome and  trends. 

Here you without programming you are making computer to learn by training the model (using known data).

Based on Model Model can take unknown data and makes prediction. For ML we have Azure Machine learning Studio 

Demo

Create a machine learning workspace. Launch ML Studio. There are three ways to learn experiment. 

Notebook: are jupiter notebook where you can write Python code. 

here you load data from data set, You  create experiment, train model and run experiment on computer code.  You see result of experiment on Experiments tab

Automated ML :  This is another way to create experiments. This help you find best matching ml alogrithm without even programming it. It's using AI to do Machine learning. Example shown if client will opt for fixed term load.

So you can load the data and based on data you get result for Model/Algorithm. Once Algorithm is selected you can deploy it as web service so you can client can send data to this end point to start getting prediction. 

Designer : You can simply drag and drop task which you normally you have to code like in flowchart format. Some task are ETL type. After pipeline is run you can view the result Score model task looks at actual and predicted price and Evaluate model task show how well the model perform on test data. Looking at Coefficient of Determination how well test data fits model say with 86% accuracy. 

You can manage the computer resources. 

Azure Cognitive services allows developer to use the existing via SDK, AI services without need to know the ML.

Azure Cognitive services can be categorized in following 

1) Computer Vision :Under Vision API there is computer vision which lets you process the images and categloge them based on things like face objects and colors or you can generate caption for images. Dev can use it even for Optical character recognition . You can read a image extract the text and return that as json. 

Video Indexer can extract audio and video insights so it's very powerful so it has privacy security implication. 

Face API, Form Api can get key value pair from documents.

Speech API : you can add speech to text or vice versa capability to your app.  There is speaker recognition 

Language API (Understand language using LUIS) Sentimental Analysis.(Analyze Text for feedback type postive or negative)

Translator Service. 

Bing Search Service. 

Decision API  Analmoly API You can enter time series data and it can detect anomalies in data. This is pattern analysis. 

Content Moderator: Found material that is offensive. . 

Personalize: Show relevant content to user. 

BOT Service 

 Created from Cognitive services , Virtual assistant for many websites uses natural language processing. 

Tools are 

Bot framework

 SDK

BOT Framework Composer ( You can configure it) 

Emulator 

Once crated you can  deploy you can deploy on azure app service and use it form there. 

 Machine learning is using existing data to forecast future behaviors, outcome and trends. 

Dev Ops Solution on Azure:

GitHub

Code Repositories: are used for working on same code, Potential of overwriting work of others. It's central space to store code. Ability to merge conflict. 

There is both Cloud and on premise version available for  Azure DevOps server GitHub Enterprise.

Azure dev ops support popular project management tool. 

Agile 

Scrum 

Basic and CMMI 

DevOps links project management with actual work. 

Philosophy of DevOps is CI (Continuous integration) and CD (Continuous delivery)

CI: is developer making continuous changes and running test with each change.  

CD: Previously we had one big chunk of deployment now it is more frequent and smaller changes. Now automation deployment script connect to environment and do the deployment.

Through Azure pipeline you achieve CI and CD.