-
### What happens?
I think we need to add in some safeguards when calculating `log2(*bayes_factors) AS match_weight`.
I didn't include a reproducible example, but I think you can see how this wou…
-
### Is your proposal related to a problem?
I would like to be able to use Splink with embedding-based similarity functions, specifically with duckdb and Athena backends.
For example, to evaluate…
-
### Is your proposal related to a problem?
Currently, when training m parameters using EM estimation, splink will not train m values for a column if any of it's comparison levels include any of the…
-
I don't this this is correct:
https://github.com/moj-analytical-services/splink/blob/42de8da5e9ce50f26c6c7de287d167c421e08dd5/splink/comparison_level_library.py#L252
Because the comparison is bein…
-
See https://github.com/moj-analytical-services/splink/discussions/769_
Hello!
I've been trying to make use of the comparing two records feature, but I keep getting the following error: does anyone…
-
I always find myself looking at both of them, and it would be nice if they were horizontally concatenated so each comparison and level lined up. This would make there be less boilerplate code needed f…
-
Hi and thanks for this amazing project!
We've recently started using Splink with Athena and it works great. But we want to move to Amazon Redshift as it's an _actual_ database. I think Redshift wou…
-
### What happens?
$m$- and $u$- probabilities are not calculated to be the same across different runs of Splink (using a `DuckDBLinker`), even when using the `seed` parameter. I checked the match w…
-
### What happens?
The break_lineage_method "delta_files" in the spark linker is missing the write path in the save command.
![image](https://user-images.githubusercontent.com/22160983/227200804-02…
-
### What happens?
If you are trying to create a `Comparison` using dictionary levels without having a `Settings` object, then such `ComparisonLevels` will not have an associated `_sql_dialect`, which…