UniversalPetrarch

UniversalPetrarch copied to clipboard

Make UniveralPetrarch pipeline compatible

3

Now that UP is a little more stable, we need to start thinking about making it usable in production pipelines. In order for people (specifically the Spanish and Arabic teams)...

ahalterman

Make a script to download all the needed models into the right places. If they're all inside the repo, the processing bash script can use relative paths instead of absolute...

ahalterman

Error when using null actor/verb codes

1

I'm trying to run Universal Petrarch with the `null_actors` and `null_verbs` options and ran into an error. Here's the traceback: ``` Traceback (most recent call last): File "tests/test_petrarch.py", line 147,...

ahalterman

Verb phrase finder doesn't look for prepositions or direct objects

6

In looking for parts of the verb phrase, UniversalPetrarch only looks for negations, compounds, and adverbial clause modifiers. It doesn't look for prepositions ("fired AT targets") or direct objects ("chanted...

ahalterman

Problem with coding source actor role and target actor

1

Consider the sentence `The United States will remove its military from Georgia.` UniversalPetrarch does not code the source actor as `USAMIL` (which is admittedly a hard one) but more importantly...

ahalterman

Function `skip_item()` in PETRgraph.py is, at best, specific to English

1

but more generally looks like a debugging function (I think it was Clayton's, not mine), as "DUCATS" were a valued item produced by those lovable bankers in the Venetian Republic...

philip-schrodt

In `AFP_SPA_19940921.0205_7.0`, adding the two actual target actors (that is, if this were annotated correctly, there should be two events based on the compound target) -CARDENAL_PAULO_EVARISTO_ARNS_ [BRXCHRARN] -LUIZ_INACIO_LULA_DA_SILVA_ [BRXOPPSIL] to...

philip-schrodt

The part of the verb matching code that seems to be for matching prepositions is not implemented. ([code](https://github.com/openeventdata/UniversalPetrarch/blob/b0cfdd045ffec285384f282a6be850c061f6c096/UniversalPetrarch/PETRgraph.py#L1209)) The preposition matching code that's in the current version is for the...

ahalterman

Add UP/dependencies white paper

7

It would be really useful as we're debugging UP on Arabic (and for users more generally) to have a white paper or algorithm describing how UP uses the dependency parse,...

ahalterman

Very high false coding rate

7

(Just making an issue for what @philip-schrodt has reported elsewhere so we can consolidate discussion): UniversalPetrarch is enormously overproducing events. Not sure what the source of the false positives is....

ahalterman