Spaces:
Running
Running
File size: 2,620 Bytes
b781b11 ec41680 b781b11 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 |
import gradio as gr
import pandas as pd
import spacy
import concepcy
LIST_RELATIONS = ["RelatedTo", "FormOf", "IsA", "PartOf", "HasA", "UserFor", "CapableOf", "AtLocation", "Causes",
"HasSubevent", "HasFirstSubevent", "HasLastSubevent", "HasPrerequisite", "HasProperty",
"MotivatedByGoal", "ObstructedBy", "Desires", "CreatedBy", "Synonym", "Antonym", "DistinctFrom",
"DerivedFrom", "SymbolOf", "DefinedAs", "MannerOf", "LocatedNear", "HasContext", "SimilarTo",
"EtymologicallyRelatedTo", "EtymologicallyDerivedFrom", "CausesDesire", "MadeOf", "ReceivesAction",
"ExternalURL"]
nlp = spacy.load("en_core_web_sm")
nlp.add_pipe(
"concepcy",
config={
"relations_of_interest": LIST_RELATIONS,
"filter_missing_text": True,
"filter_edge_weight": 2,
}
)
def greet(text, relations, to_enrich):
doc = nlp(text)
outs0, outs1 = [], []
if "document" in to_enrich:
outs0 = []
for relation in relations:
for r in doc._.get(relation.lower()).values():
outs0.extend([[relation, elt["text"]] for elt in r])
if "token" in to_enrich:
for token in doc:
for relation in relations:
rels = token._.get(relation.lower())
if len(rels) > 0:
print(rels)
print([[token.text, relation, r["text"]] for r in rels])
outs1.extend([[token.text, relation, r["text"]] for r in rels])
return pd.DataFrame(outs0, columns=["relation", "text"]), pd.DataFrame(outs1, columns=["word", "relation", "text"])
iface = gr.Interface(
fn=greet,
title="Playground for concepCy",
description="This demo enables you to play around with SpaCy's concepCy wrapper, a wrapper for ConceptNet!"
"To get started: enter a piece of text, check the relations you are interested in and if you want to "
"retrieve relations at a document-level and/or token-level.\n"
"Relations will be displayed as tables on the right hand side! Have fun!",
inputs=[
gr.Textbox(placeholder="Enter sentence here...", lines=5, value="I love eating pizzas"),
gr.CheckboxGroup(choices=LIST_RELATIONS, value=["IsA"]),
gr.CheckboxGroup(choices=["document", "token"], value=["document"])
],
outputs=[
gr.Dataframe(headers=["relation", "text"], label="Document-level relations"),
gr.Dataframe(headers=["word", "relation", "text"], label="Token-level relations")
]
)
iface.launch()
|