Katharine Xiao

Add support for composite keys

11

### Problem Description Currently, SDV only supports individual column keys as primary keys. We should extend this to support composite keys (multiple columns that uniquely identify each row when put...

Improve CTGAN performance by customizing categorical column transformations

2

### Problem Description CTGAN fails to finish in reasonable times on datasets like Airbnb and Housing Market. ### Expected behavior We should review CTGAN speed on big datasets by figuring...

Improve categorical column quality: Use label encoding (+noise)

1

### Problem Description When modeling a categorical column with two categories, sometimes the less frequent value is not sampled. ### Expected behavior Modeling the binary categorical column as a boolean...

The field distribution mapping in GaussianCopula contains internally transformed columns (e.g. `col.value` and `col.is_null`). Should the user-facing API contain those columns? Is there a way to make it more user...

Remove support for py36

1

Remove support for py36

Conditional sampling in HMA1

1

### Problem Description Add support for conditional sampling in HMA1 ### Expected behavior Users should be able to call `sample` with conditions in HMA1.

### Problem Description Currently, we are fixing the random_state using a [decorator](https://github.com/sdv-dev/Copulas/blob/master/copulas/__init__.py#L39) that wraps around the desired functions (sample). This wrapper sets the global random state to the model's desired...

### Problem Description Current, the summary metrics from the `sdgym summary` command are `% coverage, avg time, avg score`. We can update them to be more informative about the synthesizers'...

Add multi table quality report

1

Resolves #186