Create code examples for different tasks

[dvadym]

Context

Terminology can be found here.

There are 2 APIs:

1. Core APIs - which represent by public functions of DPEngine (example of usage)
2. High-level APIs: PrivatePCollection (Beam), PrivateRDD (Spark) are private counter-part of corresponding Beam/Spark objects.
   They have nice guarantee: only data that has been aggregated in a DP manner, using no more than the specified privacy budget, can be extracted (Beam example, Spark example).

It would be nice to have examples for all APIs.

Goals

There are several end-to-end examples on Github. The goal of this issue is to create examples with solving different tasks which PipelineDP user might have. Those examples might be snippets (not neccessary end-to-end). We will put them on pipelinedp.io on snippets web-page (to be created).

The initial list of those examples:

1. How to compute 1 metric DP Count.
2. How to compute 1 metric DP Sum.
3. How to compute multiple metrics per at one aggregation.
4. Using public_partitions.
5. Computing histogram with VECTOR_SUM aggregation and 1-hot encoding.
6. Computing DP aggregations for different values (i.e. by calling several times DPEngine.aggregate).

[eswarijayakumar]

@dvadym I am interested in working on this. Please assign it to me

[dvadym]

Thank you!

[system-reboot]

@dvadym is this issue still open? I would love to contribute towards it.

[dvadym]

@system-reboot yes this issue is open and it will be useful to have more examples. You're welcome to contribute.

You can start with a simple example with COUNT. Please let me know if you have any questions.