Justin Reese
Justin Reese
@TriageDr adding - if there are ingests for which publication info is particularly important, we can make a ticket to add this info ASAP
> @justaddcoffee -- thanks so much for being so open to using these mappings. Even the super gnarly ones like the one shown above, I really, really, really appreciate it!!...
Adding - this functionality is kinda already implemented: ``` from grape import datasets df = datasets.get_all_available_graphs_dataframe() df[df['name']=='KGCOVID19'] ```
TODO - deal with user supplied holdouts versus Ensmallen generated holdouts
TODO: - drop in Embiggen's new and improved BERT stuff with @LucaCappelletti94
It'd be helpful to have the YAML you are using
Also it'd be helpful to have the graphs if possible. These are likely just harmless warnings: ``` can't find key in YAML: 'pos_validation' can't find key in YAML: 'neg_validation' ```...
I think I'd need the graph files (nodes and edges tsv) to figure out what the problem is here. Seems to be an issue parsing the callbacks for MLP [here](https://github.com/Knowledge-Graph-Hub/NEAT/blob/05c2e1801944e4ef865b34fbba95131ccff55a0d/yaml_ml_instructions/IMGVR_sample.yaml#L137)....
@saadljazouli - what would be required from @LucaCappelletti94 and @zommiommy for us to update the Ensmallen version in N3C? I think just an updated conda package? Our code in N3C...
(note to self - I've already implemented this, essentially just need to expose the code to the YAML)