Hi,
Does anyone have a brief usage guide for the C deployment, basic things like what all the .exes do and the parameters you need to pass.

Also, what exe do I use to do address expansion?

Thx

The installation instructions just say to execute "npm install talisman" but don't say what is necessary after that.

On Node 8.11, "const talisman = require('talisman')" throws "Error: cannot find module 'talisman'"

The usage documentation uses the "import" feature that is not yet supported on the current LTS, so some hints about how to get going would be greatly appreciated.

The classifier LogisticRegressionClassifier.prob returns non-match probability instead of match probability. Is this dataset dependent behavior?

More thorough unit testing on feature #163

I'd love to see you record a 3 min video added as a link on your README.md might help to quickly understand the scope/applicability/benefits.

Your WHAT statement is not enough for fully understanding the WHY and BENEFIT provided.
I'd suggest to add a continued sentence, "...large-scale third-party catalogs, so that you can ..." to the
"soweego is a pipeline that connects Wikidata to large-sc

Labeled training example pairs should be stored in a table for selection and reuse. Data stored for examples should include:

• Source
• Source ids
• Label
• Label date
• Comment to store labeling rules applied by labeler

Storing examples like this allows them to by reused in the following ways:

• Select specific subsets of labeled pairs to build models from
• Store multiple labels f

Currently, the coverage report that the Azure pipeline is sending to codecov does not seem to include its origin, i.e. that it has been generated from Azure Dev-Ops and where to find information about it.

It is currently not clear what these are doing in the CLI tool, and how to interpret them. Documentation would make this more clear.

Currently only the Base64 serialised CLK format is documented. The binary one should be documented also.