Transformation Constructors#
This section gives a highlevel 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 nondifferentiallyprivate bit revealing the existence of nullity in the dataset. If we were to implicitly drop nulls from sized aggregations, then the sensitivity of nonnull individuals is underestimated. Therefore, aggregators must be fed completely nonnull data. We can ensure data is nonnull 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 nonnull 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 (
f32
andf64
).
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 















Indexing#
Indexing operations provide a way to relabel categorical data, or bin numeric data into categorical data.
These operations work with usize data types: an integral data type representing an index.
opendp.trans.make_find()
finds the index of each input datum in a set of categories.
In other words, it transforms a categorical data vector to a vector of numeric indices.
>>> finder = (
... make_find(categories=["A", "B", "C"]) >>
... # impute any input datum that are not a part of the categories list as 3
... make_impute_constant(3, DA=OptionNullDomain[AllDomain["usize"]])
... )
>>> finder(["A", "B", "C", "A", "D"])
[0, 1, 2, 0, 3]
opendp.trans.make_find_bin()
is a binning operation that transforms numerical input data to a vector of bin indices.
>>> binner = make_find_bin(edges=[1., 2., 10.])
>>> binner([0., 1., 3., 15.])
[0, 1, 2, 3]
opendp.trans.make_index()
uses each indicial input datum as an index into a category set.
>>> indexer = make_index(categories=["A", "B", "C"], null="D")
>>> indexer([0, 1, 2, 3, 2342])
['A', 'B', 'C', 'D', 'D']
You can use combinations of the indicial transformers to map hashable data to integers, bin numeric types, relabel hashable types, and label bins.
Indexer 
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 on individuallevel data.
Aggregators that produce scalarvalued statistics have a output_metric of AbsoluteDistance[TO]
.
This output metric can be chained with most noiseaddition measurements interchangeably.
However, aggregators that produce vectorvalued 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.meas.make_base_geometric()
and opendp.meas.make_base_laplace()
.
If you set the output metric to L2Distance[TOA]
, you can chain with L2 mechanisms like opendp.meas.make_base_gaussian()
.
The constructor opendp.meas.make_count_by()
does a similar aggregation as opendp.trans.make_count_by_categories
,
but does not need a category set (you instead chain with opendp.meas.make_base_ptr()
, which uses the proposetestrelease framework).
The make_sized_bounded_covariance
aggregator is Rustonly 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) 

