
This is a picture series to depict machine learning Azure algorithms & solution as part of practice. It is intended to give a quick overview of the components in Azure for machine learning solutions.
You can understand this better after going through basic machine learning development process. But if you do not, you may still give it a go.
The first goal of machine learning is to train models. So, below are some models I trained.

Creating models become easy with proper data cleaning & transformation. Below is an example.


These models were exposed as web service in azure, which allows to test any new data.

Here’s peek at data sets used.

Below are some experiments which I developed. In fact, Machine Learning modelling is possible using experiments in Azure.

It would be interesting to see what these experiments are.









The heart of Machine Learning in Azure is creating these experiments or data flows. And these allow all processing logic, like calling Python or R scripts, SQL transformation, data transformation, data format conversions, applying math or string functions, Feature Selection, Split/Join data, normalize data, editing metadata, Statistical functions, Text Analytics, Time Series & finally train, score and evaluate models based on Machine Learning Algorithms.
Finally, if you would like to get to Azure, you can register here: https://studio.azureml.net/
More articles can be found here: https://www.etechaas.com/tag/machine-learning/
Azure AzureML Data Flow Experiments Machine Learning Web Services