joshua-oss
joshua-oss
This issue is for discussion of the API for data transforms. # Overview PROBLEM STATEMENT: Various differentially private data processing routines have different expectations about the format of input data,...
Thread for synth factory design brainstorm and discussion. ### Goals Current design requires callers to know concrete synthesizer class name, import appropriate classes, and in some cases call with a...
Per: https://github.com/opendp/opendp/discussions/379 The get_query_cost method should allow an arbitrary list of queries, and return the total spend for all queries in the batch. If the caller wants to keep budget...
Issue #432 was masked by the fact that the `train` method on `DPCTGAN` and `PATECTGAN` takes an `ordinal_columns` and a `categorical_columns` parameter, but no `continuous_columns`. This is a confusing API,...
The rewriter currently generates the subquery using a Postgres-compatible AST, with the individual `XXXSerializer` classes responsible for converting the AST to the target engine's dialect. Although SQLAlchemy cannot map arbitrarily...
The rewriter currently does not apply any engine-specific name comparison rules, despite being guaranteed to have a `NameCompare` class available. The name comparison rules are currently applied only at symbol...
The SQL layer requires all aggregates to be computed precisely by the backend engine. This enables us to deal with very large datasets that cannot efficiently be stored in memory....
Currently assume USING( ) syntax for joins. Validator and adapter to support limited ON syntax for input and output. May require attention to Postgres behavior with unused aliases on subqueries
Allow queries to use ANON_* with bounds syntax for compatibility with Google Postgres extensions