Transformation Constructors#
This section gives a high-level overview of the transformations that are available in the library. Refer to the Transformation section for an explanation of what a transformation is.
As covered in the Chaining section, the intermediate domains need to match when chaining. Each transformation has a carefully chosen input domain and output domain that supports their relation.
Preprocessing is the series of transformations that shape the data into a domain that is conformable with the aggregator.
You will need to choose the proper transformations from the sections below in order to chain with the aggregator you intend to use. The sections below are in the order you would typically chain transformations together, but you may want to peek at the aggregator section at the end first, to identify the input domain that you’ll need to preprocess to.
Dataframe#
These transformations are for loading data into a dataframe and retrieving columns from a dataframe.
If you just want to load data from a CSV or TSV into a dataframe,
you’ll probably want to use opendp.trans.make_split_dataframe().
Use opendp.trans.make_select_column() to retrieve a column from the dataframe.
The other dataframe transformations are more situational.
Be warned that it is not currently possible to directly load and unload dataframes from the library in bindings languages!
You need to chain with make_select_column first.
Preprocessor |
Input Domain |
Output Domain |
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Casting#
Any time you want to convert between data types, you’ll want to use a casting transformation. In particular, in pipelines that load dataframes from CSV files, it is very common to cast from Strings to some other type.
Depending on the caster you choose, the output data may be null and you will be required to chain with an imputer.
There is an unusual caster in this section that allows you to cast from a SubstituteDistance metric to a SymmetricDistance metric.
It is noted in the linked dev issue that most transformations need only be defined for SymmetricDistance.
If you are more comfortable working with substitution distances,
you can chain this caster at the start of a computation pipeline to make d_in a substitution distance instead of a symmetric distance.
Caster |
Input Domain |
Output Domain |
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Imputation#
Null values are tricky to handle in a differentially private manner. If we were to allow aggregations to propagate null, then aggregations provide a non-differentially-private bit revealing the existence of nullity in the dataset. If we were to implicitly drop nulls from sized aggregations, then the sensitivity of non-null individuals is underestimated. Therefore, aggregators must be fed completely non-null data. We can ensure data is non-null by imputing.
When you cast with opendp.trans.make_cast() or opendp.trans.make_cast_default(),
the cast may fail, so the output domain may include null values (OptionNullDomain and InherentNullDomain).
We have provided imputation transformations to transform the data domain to the non-null VectorDomain<AllDomain<TA>>.
You may also be in a situation where you want to bypass dataframe loading and casting because you already have a vector of floats loaded into memory. In this case, you should start your chain with an imputer if the floats are potentially null.
- OptionNullDomain:
A representation of nulls using an Option type (
Option<bool>,Option<i32>, etc).- InherentNullDomain:
A representation of nulls using the data type itself (
f32andf64).
The opendp.trans.make_impute_constant() transformation supports imputing on either of these representations of nullity,
so long as you pass the DA (atomic domain) type argument.
Imputer |
Input Domain |
Output Domain |
|---|---|---|
|
|
|
|
|
|
|
|
Clamping#
Many aggregators depend on bounded data to limit the influence that perturbing an individual may have on a query.
For example, the relation downstream for the opendp.trans.make_bounded_sum() aggregator is d_out >= d_in * max(|L|, |U|).
This relation states that adding or removing d_in records may influence the sum by d_in * the greatest magnitude of a record.
Any aggregator that needs bounded data will indicate it in the function name.
In these kinds of aggregators the relations make use of the clamping bounds L and U to translate d_in to d_out.
Clamping happens after casting and imputation but before resizing. Only chain with a clamp transformation if the aggregator you intend to use needs bounded data.
Clamper |
Input Domain |
Output Domain |
|---|---|---|
|
|
|
|
|
Resizing#
Similarly to data bounds, many aggregators depend on a known dataset size in their relation as well.
For example, the relation downstream for the opendp.trans.make_sized_bounded_mean() aggregator is d_out >= d_in * (U - L) / n / 2.
Notice that any addition and removal may, in the worst case, change a record from L to U.
Such a substitution would influence the mean by (U - L) / n.
Any aggregator that needs sized data will indicate it in the function name.
In these kinds of aggregators, the relations need knowledge about the dataset size n to translate d_in to d_out.
Resizing happens after clamping. Only chain with a resize transformation if the aggregator you intend to use needs sized data.
At this time, there are two separate resize transforms: one that works on unbounded data, and one that works on bounded data. We intend to merge these in the future.
Resizer |
Input Domain |
Output Domain |
|---|---|---|
|
|
|
|
|
Aggregators#
Aggregators compute a summary statistic computed on individual-level data.
Aggregators that produce scalar-valued statistics have a output_metric of AbsoluteDistance[TO].
This output metric can be chained with most noise-addition measurements interchangeably.
However, aggregators that produce vector-valued statistics like opendp.trans.make_count_by_categories()
provide the option to choose the output metric: L1Distance[TOA] or L2Distance[TOA].
These default to L1Distance[TOA], which chains with L1 noise mechanisms like opendp.trans.make_base_geometric() and opendp.trans.make_base_laplace().
If you set the output metric to L2Distance[TOA], you can chain with L2 mechanisms like opendp.trans.make_base_gaussian().
The opendp.trans.make_count_by() transformation stands out;
it is currently the only transformation that pairs with opendp.trans.make_base_stability().
The make_sized_bounded_covariance aggregator is Rust-only at this time because data loaders for data of type Vec<(T, T)> are not implemented.
Aggregator |
Input Domain |
Output Domain |
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
make_sized_bounded_covariance (Rust only) |
|
|