Harry Caufield

Add process for generating schema-compliant data with an LLM

1

We can generate prompts from schemas, as in OntoGPT or #1499, we can generate schemas themselves, and we can populate schema entries given context _a la_ CurateGPT. Can we also...

Update to use most recent `grape` functions

1

- [x] Update run_classifier to use new `grape` methods - [ ] Remove obsoleted methods - [ ] Update tests and test fixtures - [ ] Include list of valid...

Output link predictions as SSSOM

1

The existing output for predicted links looks like this: ``` source_node destination_node score ENSP00000451575 ENSP00000435370 0.9370759965425138 ENSP00000451575 ENSP00000435370 0.9370759965425138 ENSP00000451575 ENSP00000361636 0.9361207132288921 ENSP00000451575 ENSP00000361636 0.9361207132288921 ENSP00000451575 ENSP00000357879 0.9361171909487621 ``` [The...

Drop support for Tensorflow and scikit-learn methods

1

We currently support TF and scikit-learn methods in config files, both in running classifiers and applying models. If we remove support for these two frameworks: * The neat-schema will be...

A simple ReadTheDocs would be fine for now. This may even be the same task as #71

Perhaps with Sphinx

... include line number in [schema] validation errors ... _Originally posted by @caufieldjh in https://github.com/Knowledge-Graph-Hub/neat-ml/issues/94#issuecomment-1259774884_

Replace and/or refactor functions here: https://github.com/Knowledge-Graph-Hub/neat-ml/blob/main/neat_ml/run_classifier/run_classifier.py to use the current `grape` functions. One big help would be replacing candidate pair generation with `grape`'s negative graph generation.

Updating for `grape` means we've lost some history files, though I suspect they can produced with the right parameters.

In link prediction, filter nodes by prefix or other slots

4

Some graphs have nodes we would like to filter for, but they don't make clear distinctions in their Biolink categories: ``` PR:000002977 biolink:NamedThing Graph owl:Class ``` So we would like...